Python Programming for Finance.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "collapsed": true
   },
   "source": [
    "# Python Programming for Finance\n",
    "\n",
    "This notebook includes notes, and examples from the YouTube series by [sentdex](https://www.youtube.com/channel/UCfzlCWGWYyIQ0aLC5w48gBQ), [Python Programming for Finance](https://www.youtube.com/playlist?list=PLQVvvaa0QuDcOdF96TBtRtuQksErCEBYZ).\n",
    "\n",
    "## Required Packages\n",
    "\n",
    "* pandas\n",
    "* pandas-datareader\n",
    "* matplotlib\n",
    "* beautifulsoup4\n",
    "* scikit-learn\n",
    "* mpl_finance\n",
    "\n",
    "\n",
    "## Table of Contents\n",
    "\n",
    "#### Section 1 - Webscraping\n",
    "\n",
    "1. [Analyse data from just one company](#intro)\n",
    "\n",
    "2. [Collect and combine data from the entire S&P 500 companies](#SP)\n",
    "\n",
    "3. [Analyse relationships between companies](#corr)\n",
    "\n",
    "4. [Formulate trading strategies](#form_strat)\n",
    "\n",
    "5. [Analyze trading strategies](#an_strat)\n",
    "\n",
    "_________________________\n",
    "\n",
    "#### Section 2 - Quantopian\n",
    "\n",
    "1. [Algorithmic trading](#quant)\n",
    "2. [Orders and leverage](#orders)\n",
    "3. [Schedule function](#sched)\n",
    "4. [Research](#res)\n",
    "5. [Pipeline](#pipe)\n",
    "6. [Analyzing Alpha Factors with Alphalens](#alpha)\n",
    "7. [Backtesting Alpha Factor](#backtest_alpha)\n",
    "8. [Analyzing Backtest with Pyfolio](#pyfolio)\n",
    "9. [Strategizing](#strategizing)\n",
    "10. [Finding more Alpha Factors](#findalpha)\n",
    "11. [Combing Alphas](#combinealpha)\n",
    "12. [Portfolio Optimization API](#portopt)\n",
    "\n",
    "_________________________\n",
    "\n",
    "#### Section 3 - Zipline\n",
    "\n",
    "1. [Local Backtesting and Finance with Python](#localback)\n",
    "2. [Visualizing Strategy Metrics](#visstrat)\n",
    "3. [Custom Data Panel](#customdat)\n",
    "4. [Trading Custom Markets (Bitcoin example)](#tradecustom)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a",
    "collapsed": true
   },
   "source": [
    "## <a class=\"anchor\" id=\"intro\"></a> Intro and Getting Share Price Data \n",
    "\n",
    "To begin, we'll collect and analyse the share price data of the electric car company [TESLA](https://www.tesla.com/en_GB/). This was done using the pandas DataReader, and by collecting the data from the financial website https://www.morningstar.com/. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Close</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Open</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-06-29</th>\n",
       "      <td>23.89</td>\n",
       "      <td>25.0000</td>\n",
       "      <td>17.54</td>\n",
       "      <td>19.00</td>\n",
       "      <td>18783276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-06-30</th>\n",
       "      <td>23.83</td>\n",
       "      <td>30.4192</td>\n",
       "      <td>23.30</td>\n",
       "      <td>25.96</td>\n",
       "      <td>17194394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <td>21.96</td>\n",
       "      <td>25.9200</td>\n",
       "      <td>20.27</td>\n",
       "      <td>25.00</td>\n",
       "      <td>8216789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-02</th>\n",
       "      <td>19.20</td>\n",
       "      <td>23.1000</td>\n",
       "      <td>18.71</td>\n",
       "      <td>23.00</td>\n",
       "      <td>5135795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-05</th>\n",
       "      <td>19.20</td>\n",
       "      <td>19.2000</td>\n",
       "      <td>19.20</td>\n",
       "      <td>19.20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Close     High    Low   Open    Volume\n",
       "Date                                              \n",
       "2010-06-29  23.89  25.0000  17.54  19.00  18783276\n",
       "2010-06-30  23.83  30.4192  23.30  25.96  17194394\n",
       "2010-07-01  21.96  25.9200  20.27  25.00   8216789\n",
       "2010-07-02  19.20  23.1000  18.71  23.00   5135795\n",
       "2010-07-05  19.20  19.2000  19.20  19.20         0"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')  # Hide warnings\n",
    "import datetime as dt\n",
    "import pandas as pd\n",
    "pd.core.common.is_list_like = pd.api.types.is_list_like\n",
    "import pandas_datareader.data as web\n",
    "import numpy as np\n",
    "\n",
    "start = dt.datetime(2000, 1, 1)\n",
    "end = dt.datetime.now()\n",
    "df = web.DataReader(\"TSLA\", 'morningstar', start, end)  # Collects data\n",
    "df.reset_index(inplace=True)\n",
    "df.set_index(\"Date\", inplace=True)\n",
    "df = df.drop(\"Symbol\", axis=1)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Close</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Open</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-07-23</th>\n",
       "      <td>303.20</td>\n",
       "      <td>305.5000</td>\n",
       "      <td>292.8601</td>\n",
       "      <td>301.8400</td>\n",
       "      <td>10992947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-24</th>\n",
       "      <td>297.43</td>\n",
       "      <td>307.7171</td>\n",
       "      <td>292.5452</td>\n",
       "      <td>304.4209</td>\n",
       "      <td>9590784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-25</th>\n",
       "      <td>308.74</td>\n",
       "      <td>309.6200</td>\n",
       "      <td>294.5025</td>\n",
       "      <td>296.7400</td>\n",
       "      <td>7094180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-26</th>\n",
       "      <td>306.65</td>\n",
       "      <td>310.7000</td>\n",
       "      <td>303.6411</td>\n",
       "      <td>304.8500</td>\n",
       "      <td>4630522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-27</th>\n",
       "      <td>297.18</td>\n",
       "      <td>307.6933</td>\n",
       "      <td>295.3404</td>\n",
       "      <td>307.2500</td>\n",
       "      <td>5703326</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Close      High       Low      Open    Volume\n",
       "Date                                                      \n",
       "2018-07-23  303.20  305.5000  292.8601  301.8400  10992947\n",
       "2018-07-24  297.43  307.7171  292.5452  304.4209   9590784\n",
       "2018-07-25  308.74  309.6200  294.5025  296.7400   7094180\n",
       "2018-07-26  306.65  310.7000  303.6411  304.8500   4630522\n",
       "2018-07-27  297.18  307.6933  295.3404  307.2500   5703326"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize TESLA Close Prices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the close prices\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "\n",
    "style.use('ggplot')\n",
    "\n",
    "df.Close.plot()\n",
    "plt.ylabel('Close Price'); plt.xlabel('Date')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Basic Stock Data Manipulation\n",
    "\n",
    "### Moving Averages\n",
    "\n",
    "Here we will create [moving averages](https://www.investopedia.com/terms/m/movingaverage.asp) of the TESLA share price data. \n",
    "\n",
    "Moving averages are created by averaging over historical price data, and result in a curve that smooths out price over time. They can be useful indicators for traders, particuarly when different moving averages cross as this can indicate an uptrend/ downtrend in price. An uptrend commonly occurs when a shorter day (low latency) moving average crosses and remains above a longer day (high latency) moving average, and vice versa for a downtrend. \n",
    "\n",
    "Here we will create and visualize the 100 day moving average, and visualize it along with the close price and volume data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Close</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Open</th>\n",
       "      <th>Volume</th>\n",
       "      <th>100ma</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-06-29</th>\n",
       "      <td>23.89</td>\n",
       "      <td>25.0000</td>\n",
       "      <td>17.54</td>\n",
       "      <td>19.00</td>\n",
       "      <td>18783276</td>\n",
       "      <td>23.890000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-06-30</th>\n",
       "      <td>23.83</td>\n",
       "      <td>30.4192</td>\n",
       "      <td>23.30</td>\n",
       "      <td>25.96</td>\n",
       "      <td>17194394</td>\n",
       "      <td>23.860000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <td>21.96</td>\n",
       "      <td>25.9200</td>\n",
       "      <td>20.27</td>\n",
       "      <td>25.00</td>\n",
       "      <td>8216789</td>\n",
       "      <td>23.226667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-02</th>\n",
       "      <td>19.20</td>\n",
       "      <td>23.1000</td>\n",
       "      <td>18.71</td>\n",
       "      <td>23.00</td>\n",
       "      <td>5135795</td>\n",
       "      <td>22.220000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-05</th>\n",
       "      <td>19.20</td>\n",
       "      <td>19.2000</td>\n",
       "      <td>19.20</td>\n",
       "      <td>19.20</td>\n",
       "      <td>0</td>\n",
       "      <td>21.616000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Close     High    Low   Open    Volume      100ma\n",
       "Date                                                         \n",
       "2010-06-29  23.89  25.0000  17.54  19.00  18783276  23.890000\n",
       "2010-06-30  23.83  30.4192  23.30  25.96  17194394  23.860000\n",
       "2010-07-01  21.96  25.9200  20.27  25.00   8216789  23.226667\n",
       "2010-07-02  19.20  23.1000  18.71  23.00   5135795  22.220000\n",
       "2010-07-05  19.20  19.2000  19.20  19.20         0  21.616000"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create, and add 100 day moving average to dataframe\n",
    "df['100ma'] = df['Close'].rolling(window=100, min_periods=0).mean()\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the close prices, 100ma, and volume\n",
    "ax1 = plt.subplot2grid((6,1), (0,0), rowspan=5, colspan=1)\n",
    "ax2 = plt.subplot2grid((6,1), (5,0), rowspan=1, colspan=1, sharex=ax1)\n",
    "\n",
    "ax1.plot(df.index, df.Close)\n",
    "ax1.plot(df.index, df['100ma'])\n",
    "ax2.bar(df.index, df.Volume, label='volume')\n",
    "\n",
    "ax1.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Candlestick Charting\n",
    "\n",
    "Resampling is a very useful operation in Pandas for manipulating time series data. It involves changing the frequency of your time series observations. \n",
    "\n",
    "There are two types of resampling:\n",
    "1. ___Upsampling___: Where you increase the frequency of the samples (e.g. minutes $\\rightarrow$ seconds)\n",
    "2. ___Downsampling___: Where you decrease the frequency of the samples (e.g. days $\\rightarrow$ months)\n",
    "\n",
    "Here we will use resampling to create [candlesticks](https://www.investopedia.com/trading/candlestick-charting-what-is-it/). \n",
    "\n",
    "First, we resample our close prices to get the [open-high-low-close](https://www.investopedia.com/terms/o/ohlcchart.asp) (OHLC) data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>open</th>\n",
       "      <th>high</th>\n",
       "      <th>low</th>\n",
       "      <th>close</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-06-29</th>\n",
       "      <td>23.89</td>\n",
       "      <td>23.89</td>\n",
       "      <td>15.80</td>\n",
       "      <td>17.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-09</th>\n",
       "      <td>17.40</td>\n",
       "      <td>20.64</td>\n",
       "      <td>17.05</td>\n",
       "      <td>20.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-19</th>\n",
       "      <td>21.91</td>\n",
       "      <td>21.91</td>\n",
       "      <td>20.22</td>\n",
       "      <td>20.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-29</th>\n",
       "      <td>20.35</td>\n",
       "      <td>21.95</td>\n",
       "      <td>19.59</td>\n",
       "      <td>19.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-08-08</th>\n",
       "      <td>19.60</td>\n",
       "      <td>19.60</td>\n",
       "      <td>17.60</td>\n",
       "      <td>19.15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             open   high    low  close\n",
       "Date                                  \n",
       "2010-06-29  23.89  23.89  15.80  17.46\n",
       "2010-07-09  17.40  20.64  17.05  20.64\n",
       "2010-07-19  21.91  21.91  20.22  20.72\n",
       "2010-07-29  20.35  21.95  19.59  19.59\n",
       "2010-08-08  19.60  19.60  17.60  19.15"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Resample to get open-high-low-close (OHLC) on every 10 days of data\n",
    "df_ohlc = df.Close.resample('10D').ohlc() \n",
    "df_volume = df.Volume.resample('10D').sum()\n",
    "\n",
    "df_ohlc.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```matplotlib.finance``` offers some useful functions for plotting candlesticks. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mpl_finance import candlestick_ohlc\n",
    "import matplotlib.dates as mdates"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The function ```candlestick_ohlc``` requires a special type of data type known as an mdate. Here, we convert our dates into mdates:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>open</th>\n",
       "      <th>high</th>\n",
       "      <th>low</th>\n",
       "      <th>close</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>733952.0</td>\n",
       "      <td>23.89</td>\n",
       "      <td>23.89</td>\n",
       "      <td>15.80</td>\n",
       "      <td>17.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>733962.0</td>\n",
       "      <td>17.40</td>\n",
       "      <td>20.64</td>\n",
       "      <td>17.05</td>\n",
       "      <td>20.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>733972.0</td>\n",
       "      <td>21.91</td>\n",
       "      <td>21.91</td>\n",
       "      <td>20.22</td>\n",
       "      <td>20.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>733982.0</td>\n",
       "      <td>20.35</td>\n",
       "      <td>21.95</td>\n",
       "      <td>19.59</td>\n",
       "      <td>19.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>733992.0</td>\n",
       "      <td>19.60</td>\n",
       "      <td>19.60</td>\n",
       "      <td>17.60</td>\n",
       "      <td>19.15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Date   open   high    low  close\n",
       "0  733952.0  23.89  23.89  15.80  17.46\n",
       "1  733962.0  17.40  20.64  17.05  20.64\n",
       "2  733972.0  21.91  21.91  20.22  20.72\n",
       "3  733982.0  20.35  21.95  19.59  19.59\n",
       "4  733992.0  19.60  19.60  17.60  19.15"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ohlc.reset_index(inplace=True)\n",
    "df_ohlc.Date = df_ohlc.Date.map(mdates.date2num)\n",
    "df_ohlc.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Having created our open high low close data, we can now use it to create candlesticks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJztnXt8XGWd/9/TJE0bGkrbtJQ0s5RSQKtCqa61yA/vLFZZ3F9/PlF3BRXF34qrstoFye5PV816qYq4AmsRBFaFeaCy3CqWeymUQBtber+ll5mmbZo0TdMmTSYzz++Pc2Yyk8zlzGQuJ5Pv+/XKK3Oe85xzPjOTfM/3fJ/v83w9xhgEQRCE0mVcsQUIgiAI+UUMvSAIQokjhl4QBKHEEUMvCIJQ4oihFwRBKHHE0AuCIJQ4YugFQRBKnHKnHZVSZcA64KDW+uNKqfOAh4BpwHrgs1rrfqVUJfAA8E6gA6jXWu/LuXJBEATBEZl49F8HtsVs/xi4TWs9F+gErrfbrwc67fbb7H6CIAhCkXBk6JVSdcDHgN/Y2x7gg8Ajdpf7gU/Yr6+xt7H3f8junwojP/IjP/IjP1n9pMVp6OYXwL8A1fb2NOC41nrA3g4As+zXswA/gNZ6QCnVZfdvT3WB1tZWh1LyS01NDe3tKaUWHDdqAtGVCW7UBKIrE9yoqba21lG/tIZeKfVxoE1rvV4p9f4R6oo97w3ADQBaa2pqanJ16hFRXl7uGi0R3KgJRFcmuFETiK5McKMmpzjx6N8L/K1SajEwATgTuB04SylVbnv1dcBBu/9BwAsElFLlwGSsQdk4tNbLgeX2pnHLndKNd203agLRlQlu1ASiKxPcqMmpR582Rq+1/rbWuk5rPRv4FPC81vrvgReA/2N3uw54zH79uL2Nvf95rbWjOJIgCIKQe0aSR38z8M9Kqd1YMfh77PZ7gGl2+z8Dt4xMoiAIgjASHOfRA2itXwRetF+3AO9O0Oc08MkcaBMEQRBygMyMFQRBKHHE0AuCIJQ4YugFQShJJvp8afv4dqbvUwqIoRcEoeTw7fRxxr33Mnnp0pT9At2BAikqLmLoBUEoOQLdAcZ1dlLe0lJsKa5ADL0gCCXHqv2r6OzrLLYM1yCGXhCEkqPzdCcD4YG0/fzdfqobGwugqLhklEcvCIJQSvi7/Yxv9hdbRt4Rj14QhJLCzZk0k5cuZdqSJQW/rhh6QRBKCjdn0pS3tFDmL/wThBh6QRDGDG729vOJGHpBEMYMge5A3ESqBTMWFFFN4RBDLwhCSXJoaiUDc+YMay8PDIZ2GhY2FFJS0XBSYWoCsBqotPs/orX+jlLqPuB9QJfd9XNa6w12fdjbgcVAj93enA/xgiAIifBWe/nZ9+ew7Ipljo+Z6PPRW1+fN01LVy9l3pQt3Ow/M2/XSIYTj74P+KDW+hJgPnCVUuo99r6lWuv59s8Gu+2jwAX2zw3AXbkWLQiCkAx/t58VV69IaOQrn1lF5apVAHH5876dvjhP3ymNTc5z8Fu6WjgwsS/ja+QCJxWmjNb6pL1ZYf+kqhh1DfCAfdxrWCUHzxm5VEEQhPT4uxNntSxdvZSuQy2M67RmzI5vHgw0BLoDGWXDROL8zW2jI1jhKEavlCpTSm0A2oBntNZN9q5GpdSbSqnblFKVdtssIPYTC9htgiAIBWXakiXRhc1aulqis2Uz8cSHMtGXnfdfTBzNjNVah4D5SqmzgEeVUm8Hvg0cBsZjFfm+Gfie0wsrpW7ACu2gtXZNdXU3Vnp3oyYQXZngRk1QmroqKiqix1a0tlJRUUFFTQ0VFRXggbKyMjZ1bor2K6+ooKqqivFXXpnymhFNZceOQVsblfY5Z9x2G2EHyyjEXn/GypWEr702q/eXDZmWEjyulHoBuEpr/VO7uU8p9VvgW/b2QcAbc1id3Tb0XMuxbhAAxi3V1d1Y6d2NmkB0ZYIbNUFp6goGg9FjZ4RChIJBOtrbCQaDdFWV019bSzAYJBiEjvZ2pgWD9PT00LZ4MaS4Zk1NDTc9dRM/6pnA+N27o+cMrVlDhwOtwWAQDIRCIXq3baM7B597bW2to35pQzdKqem2J49SaiLwEWB7JO5uZ9l8AthsH/I4cK1SymMP2nZprQ9l/hYEQRByy5qLp9CxYkXS/ekWOGtua47G8mP7OilyUkycxOjPAV5QSr0JvIEVo38S+L1SahOwCagBfmD3Xwm0ALuBu4Gv5Fy1IAhCBkxeupSrH9sS19a/YHCyVGQAN3aANkKy2bSxfTOJ2Y/r7Ixm/hSKtKEbrfWbwKUJ2j+YpL8Bbhy5NEEQhMxINsha3tLCOeH41MbuBmuyVMjrZfrE6UnPOXTtnJDXm/V6NV1nWCY3kvlTKGSZYkEQSobmtmZHyxrE9ulbtIiGhcknSiVL18yGNRdPIfx8zk7nGFkCQRCEksLJsgaxfdLNho019N5qLwN1ddmLKxLi0QuCIDhkUe0iet9fT+XatQAsmDEdcP+kKfHoBUEQYkiUQRPJsKm/0PL++xYtIuT1ZrUomqmsTN8px4ihFwShpCnfsoXyPXuc9x+SQbNgxoJh2Tix4Z6Q10smhGfMyKh/LhBDLwhCyRKeMgXKRxahTue19y1aVJSqUZkghl4QhFFPsrTKviuvpOfzn2fg/PPzdu3e+nox9IIgCPnEt9OXdhXJgTlzODSt8LHxVEz0+dLOxM0VYugFQRjVOCkG3rVsGU9c87a0/foXLEjqnT/w5gMpj8uU8kAg4UzcfCCGXhCEUU26CU2RvPc5k+fgrU49cNrd0BA19EPDQfu79qc8LhVLnljCnq74AeFChnskj14QhFFNxNBXNzbCxcP3RzJkMikrODQcFPJ62Xd8X8pjUpUijGiMvdEU0tCLRy8IQkmQyzDI2lZrQlTEqw95vSk9eiA6iSoZ508+nxVXJ185M5+IoRcEYVQSO7EpXUgmUyIeeMSrH6irY+GshSmPcXPmjRh6QRBGJbETm7zV3oQTl0ayLk3swme99fU0fiA+Zp/tuQfmzCHk9WY1gJstaWP0SqkJwGqg0u7/iNb6O0qp84CHgGnAeuCzWut+u3bsA8A7gQ6gXmu9L0/6BUEY40QMsmXo473qdAuWpaJhYQPfePEbSfc7OXeiuH3XssGxgrO+kfz8ucSJR98HfFBrfQkwH7jKrhz1Y+A2rfVcoBO43u5/PdBpt99m9xMEQcg5vp2+6MzVXK0qGftkkGlIyLfTFxdSSleQJNPlE7IlraHXWhut9Ul7s8L+McAHgUfs9vuxygkCXGNvY+//kF1uUBAEIafE5tD31tc7Wos+HSGvN3qeuurMbh6B7kDUuC95Ygntf/wtk5cuHbGmkeIovVIpVYYVnpkL3AHsAY5rrQfsLgFglv16Fvbzk9Z6QCnVhRXeaR9yzhuAG+x+rqlEP5Lq8/nCjZpAdGWCGzXB6NZVVlXFW885l5qaGqqqqqipqeG2j9024muXVVVx28d+CMCNNYPF8tJpKq+ooKqqiqoq8LWupLWnlQknTjHR76fsXWVUVFQMO76sqorKAnz+jgy91joEzLeLhD8KvGWkF9ZaLweW25vGLZXoR1J9Pl+4UROIrkxwoyYY3brO2rGDxbX/SHt7Oz09PTl7HxOnTqU3wbnSaZoWDLLjyA56erxsO7SNUCgEQDAYJBQKEQwGhx2f7FpOqa2tddQvo6wbrfVx4AVgEXCWUipyo6gDDtqvDwJeAHv/ZKxBWUEQhJwRmj5Y5zXTEEsqRjKAm2nZwZFcKxPSGnql1HTbk0cpNRH4CLANy+D/H7vbdcBj9uvH7W3s/c/bBcMFQRByRuyyA5GCIG5iyoQplI9zx+IDTjz6c4AXlFJvAm8Az2itnwRuBv5ZKbUbKwZ/j93/HmCa3f7PwC25ly0IguBurjz3SqZUTim2DMBBjF5r/SZwaYL2FuDdCdpPA5/MiTpBEIRRhjVDtjBpk06RmbGCkCG+ncNrigpChNixA7cghl4QMsTJ+ufC2CUydlBXXUfI6+W++UUWhBh6QciKQlUGSoY8Vbif+gvr6Vixgp2LLyu2FDH0gpANhaoMNJTGpkaqGxvH3FNFsW+smZLLdM9cIIZeEDIk01zpXNLc1ly0m0wxKTt6NG47WTFwtxCb7rlq/yr6w/3MmTynaHrckeQpCKOIYhr6scrQxb/SFQN3E5MqJnHxX12cUYWrXCOGXhCEUUF1Y2Pa2qxupFhVpWKR0I0gCKOCsRiyyhVi6AVhFBO79nmp0tjUmLO15scqYugFYRQRKYTh7/Yz0edLW9iiFGhuay7Y4l8jJde1a3OFGHpByJJieNOxhj6ZkZ+2ZElOi11Izr5z3GroZTBWEBwSSemzqg81F9yb9u30RfOzF8xYADsT9/knf26zgtyWsz/R52PB7MIV1i4FxKMXBIc0tzVztPdotEZpmW1QCzWZZ23rWuovrCfk9cZpiH2yuHfzvXT2dRZET7EoDwSi719wRlqPXinlBR4AzsaqFbtca327Uuq7wJeAyEyGW7XWK+1jvo1VJDwEfE1r/ec8aBeEghP7aB4x9LHZIEueWMLnNsDV/5b7lLpI/n5sTnmZ3095zHbn6U4GwgPDjs3Fdd1CWY6fWHJF/wL3PmU48egHgG9qrecB7wFuVErNs/fdprWeb/9EjPw84FPA24CrgDvtmrOCUFTyEWseOkPT3+3PiyGK1V7oDBR/t79oM1FH0/iAm3P80xp6rfUhrXWz/bobq7rUrBSHXAM8pLXu01rvBXaTYN16QSgkjU0jXx/GW+2Nxsgj3luiGZqXv9mZ08FQiI+TRzJQCmnwm9uai7LeTKLvbOgsWTfhtjVuImQ0GKuUmo1VhKQJeC/wVaXUtcA6LK+/E+sm8FrMYQES3BiUUjcANwBorV1Tid5J9flC40ZNMLp0bercxPvOfd+I9F509kXceNmN1sZtt1H+4Q9TUQEVFUTPW1ZWxuRTA0z0+6mIudZIPqsH3nyAtv42Lp99efw5bryR8kcfpbyqikq7fc7UOVSUHaeiosLR9Zzoqqio4NzJ51J14kT0Ovkmoquqqir6vsvmzmXGbbfBW99aMB2JNKXixpobC6QmMxwbeqXUJGAF8A2t9Qml1F3A97Hi9t8HfgZ8wen5tNbLgeX2pnFLJXon1ecLjRs1wejSFQwG6enpyVpvY1Mjc6fMjTt+WjBIMAjBIHTY7aFQCIDw9u2Y972PjhUrkmpyet0J5RPo6+vjpkU3DTvHtGCQ/p4ebnnqJgAeuuohzrztb+j1eulycD0nuoLBIAumLaBnRgBuuqkgIYqIrqllU6l/az3t7e1Uz5jB+DVr6LjpJijC350b/95ra2sd9XNk6JVSFVhG/vda6z8CaK2PxOy/G3jS3jxIfB2tOrtNEIrKSAYVm9uah2V69C9YgLf6KBB/3q4zyplxOjeDhs1tzaiLVNqQQGwI6dQXHPtbjqm/sJ6BOh9VWuf83OmuG2Ggro4yF4dt3EzaGL1SyoNV8Hub1vrnMe3nxHT7O2Cz/fpx4FNKqUql1HnABcDruZMsCNkxfWJuS7x1NzQknCCz5uIpDJx//ojP39jUyIIZC6i/sD7O4KWjt74+Lzn+hZ6dOnQAuLe+nr5FiwqqoVRw4tG/F/gssEkptcFuuxX4tFJqPlboZh/wZQCt9RallAa2YmXs3Ki1DuVauCBkSsPCBib6fHkzWJEMkUPTKqFj5Odrbmt2xcqHxSLRQPdoWQrBbaQ19FrrNYAnwa6VKY5pBNxdGUAYk1SuXevIWDQ2NfKjZ9OnzMWGVCKTlVZedjH/kmDWar6xZuxaVK5axbjDh+lalnoN9HEPPACLF+dbmlBkZGasMKZwGjd3WskpNqSSj8lK6ehfsCD6nmLHEMZ1dlLe0pLwmMhM2samRjwvvZTy/MUs4efWdWNGI2LohTFFyOvNi/Eq1oSi7oaGjAd9ywMBfDt9NLc149m/P2XfdDe7RAu75WqS06JaicfnCjH0wpgi5PXmpYBFrkvb5XpG6NDzBboDaT1mJzevyrVrh7XlahG0TAaghdSIoRfGFJHZpKkMaabeeexMzcqyypwUgQ50B+Ji7qlwssZKxPg2NjVS5vfj7/anNfTNbc2EvN6U/YY+Tfh2+li1fxVLV49sZnDDC+5dTmA0IoZeGFNEBmJTeZ2Zeuexhv5t097GsiuWMTBnTlz7uAceyFApjldoTDVgPDS00tzWHDX0TqbrpzP0Q1nbann4V7yUeHzAKU0Hm0Z0vBCPGHpBSEDEm3ZaXGSo9921bFl0ViyQNhY+lFytGFkeCLB09VJW7V8V1+6t9joOjSS6IaR6IvrC27/AZzu8jH/11czECnlDDL0gJCDiTZcHAtGJS8kYqKtL6H1PXrqUaUuWZHX9XC4N3NLVQufpwTXqrRm9zr30RDeEta1rhw1q+3b6ojeQvkWLqP7Zz3K+uJuQHWLoBSEJsStUpgqjRMJB3mpvXHy+vKWFp/veZMkTS2Dfvrhj8l2GMOT1MjAnfqwgMvaQyVo1yVbI9Hf7KTt6NK4t0QqbgjsQQy+MaVIZ3EjqotNB0RVXr2DZFfETlI5VDODv9mNmzoxrT5StEiHdE0QihvbvWLFi2GSp5rbmjHPTUxnsyHIEkc9waNw/ErrK9KlmNK1BP1oQQy+MWaobGykPBFIa+zK/Pydl68JDwhxl/uTFPNI9QSTCSX9vtTeam56LddN7662yhpG8fH+3f1iYp7ylJeM8/0B3gIWzFo5YnzCIGHphTOLv9kfz6VMtAJbP8nC5zr0fyuSlS6lctYrFr7YB8QOw2eaoDy360bdoEWV+P2tb12b8FJKKxg/ICiq5RAy9MOYIeb0JBzsThQzcXB7OCWbSJOadqmLKhCk5Od9QQ99bXx9N13TyVJEuLNPY1OjaKk2jGTH0wpgjWSm6RJkkuSRy7lQTsnK5lELXMiuf/+odcOW5VybsE164MKOB4cjgbDbe+9LVS7l3870p+zS3NcuM2DyQdvVKpZQXeAA4G2tJ4uVa69uVUlMBHzAba5lipbXutNevvx1YDPQAn4vUnBUENzBQV2cbqvg/SyuTJH/XjYSKYkM21Y2NcU8N2QyYpqK8pYVxnZ1JveRwYyOV//APCQddE+nora+nurGRhgyfdJauXsozB57h7Kqzh71nIf848egHsOrBzgPeA9yolJoH3AI8p7W+AHjO3gb4KFaxkQuwasLelXPVgjACeuvr48IMsYOFmc4ETcXAnDnW2vRDiD1/onV3crmYV2SGbqbFS3w7fUl1ZLNWUEtXC+PHjefKc6/My1pDQmrSGnqt9aGIR6617ga2YRX7vga43+52P/AJ+/U1wANaa6O1fg04a0g1KkEoOMmMd5nfH2foB+rqcmbou5YtY8u1V0fj45GQkbfaS/+CBQnDH5GKUrli6AzdRCRa0XNt69qkOrIZoI4MCEdobGrM+1wCYZCMYvRKqdnApUATcLbW+pC96zBWaAesm0DsSFfAbhOEopHK0MfSW1+f08HAZVcsi8bH+xYtipu01LCwIWpkI+25SOXMlIG6umFedqqZudmEXeadqsJb7Y1+ts1tzSnnEgi5xVFxcACl1CSsAuHf0FqfUEpF92mtjVLKZHJhpdQNWKEdtNbU1NRkcnjeKC8vd42WCG7UBKNHV8MLDVRVVQ3TWjZ3LmbmTMY1NVFTU0NFRQU1NTXcWHNjTvVUVVVRXl7OGTfeyKbffZhrL742qqXsoovwvPQSm95JUT7LiK7yRx+Nu37ks3B0jooKLp+9MGH/8ooKPGVlfOKtn+Dqz/+b3fYoFRUw4fDhYd/T3Jq51NTUuPJvy42anOLI0CulKrCM/O+11n+0m48opc7RWh+yQzORZ7ODQKz7VGe3xaG1Xg4stzdNe3t7NvpzTk1NDW7REsGNmmD06NrfsR9vtXeY1uoZM+i+6SamLVlCR3s7MyfMzMv7mVo2lYGBAdrb2wkGgyyuXRy9zsSpU6kKBgkGKcpnGfmspgWD9NxxR3RQNhgMOtYzLRjkpnfclLD/ZK+XivZ2Bnbs4Li9v/od7yAYbCYYhI7I5+DzsaZqDeoiRXt7uyv/ttyoqba21lG/tKEbO4vmHmCb1vrnMbseB66zX18HPBbTfq1SyqOUeg/QFRPiEYSCs6h2UcJwzNB1XPJV0ShVzD1iWBfMWFDUsn2QeuJYtnQtW0b7s89Gl0sAK/QzdHXQyLUltTI/OPHo3wt8FtiklNpgt90K/AjQSqnrgf1AJJazEiu1cjdWeuXnc6pYGJM4LdadiGTGIzalsLGpsSDx8WT55w0LGxj/k+xWuswHmQxIOxmcHZq+aX3WS6Krgy7zH4ULM1UpOCWtoddarwE8SXZ/KEF/A+Q2yCmMeaxi3fk9fyFIdTNJNpGrEIS83ujAdGNTY0aGfiQ58WV+P81tfspytyqzkACZGSsIRSbiERfT0EfWrIHC3fSGPgnkcqKYEI8YeqFkWbp6qbUWvMuJeMTJ1n4vBENDK4VYb6a7oSEudp+vMRJBDL0wCohd/yWTtWBaulocV2rK5cqL2eKmYh2FGhSNfc8yEJs/xNALrqe5rTlqiDMJK7T1tKXvZFOMiUqCxYIZC/K6HLSQwYQpQSgmkSwNp5730tVLOdxzmCmV6ZfnFSNTXBoWNtAtdUbyinj0wqgi4nmnWyelpauFgfCAo3PKSopCqSOGXhiVOJncUz6uXDI5BAEx9MIoIpJ+6GQGaVtPG1Mqp7Di6tQrNwrCWEAMveBaGpsa44x6JBUv3Xrm/7jyH+ns6xRvXhBsZDBWcC3R2bAXW9tO0w93HdtFVXmVePNZ0NjU6IpUUyG3iKEXXI0VrpH58YWiua1ZbpAliIRuBFcT8nrFwywQkmZauoihF1zNQF1dwslMQytDCSNH0kxLl7ShG6XUvcDHgTat9dvttu8CXwKO2t1u1VqvtPd9G7geCAFf01r/OQ+6hTFCorh8/4IFUmBaEDLAiUd/H3BVgvbbtNbz7Z+IkZ8HfAp4m33MnUqpslyJFcYWyUI2yTxP304pNi0IiUhr6LXWq4FjDs93DfCQ1rpPa70Xq/jIu0egTxiDZFsoO9Cd+wpJglAKjCTr5qtKqWuBdcA3tdadwCzgtZg+AbtNEByTzXroE30+VlWs4n2/+RO8T5LJBCGWbP8j7gK+Dxj798+AL2RyAqXUDcANAFpr11RXd2OldzdqgvzpqqioYGXrSq69+Nrk166oiLt22bFjnJh6gg7AM2U2Zce7XPWZjZbvsGLI51os3Ph5uVGTU7Iy9FrrI5HXSqm7gSftzYNA7HTEOrst0TmWA8vtTeOW6upurPTuRk2QP13BYJBth7bRXpv83NOCQTpirl3d00NocojAWZXMnTKXY6eOueozGy3f4cwJM12h042flxs11dbWOuqXVXqlUuqcmM2/Azbbrx8HPqWUqlRKnQdcALyezTUEIR2R5RFiB2FXXjaDuxbfxZXnXlksWaMaWTaiNHGSXvkg8H6gRikVAL4DvF8pNR8rdLMP+DKA1nqLUkoDW4EB4EatdSg/0oWxTiTFMtAdoHLVKmYv7ufcujlFVjW6KUQJQaHwpDX0WutPJ2i+J0X/RsB5vTdBGCGr9q/i+tAR/hz4CF1fX1ZsOaMaKedXmsjMWMF1eKu9aT3L2On6nac7uf7r59O1bNDIi2cqCIOIoRdcRWNTI95qb1rPMt10ffFMBWEQMfSCq2huaxZvXBByjBh6wXWINy4IuUUMveAKIsseCIKQe8TQC66gua3ZUS3YWEJer9wgBMEBYugF15Dp0sMhr5fmtma81V7mTJb8eUFIhqz+JLgGq5rUdMf9B+qsQVspfScIqRGPXnANIa83o6WJe+vrpcygIDhADL1QdEYSZ890zXpBGIuIoReKTnNbs3jmgpBHxNALrkA8c0HIH2LohaIg9V0FoXA4Wab4XuDjQJvW+u1221TAB8zGWqZYaa07lVIe4HZgMdADfE5rnXldOKHkSVTfNZJFIwhCbnHi0d8HXDWk7RbgOa31BcBz9jbAR7GKjVyAVSbwrtzIFMYCvfWy9IEg5IO0hl5rvRo4NqT5GuB++/X9wCdi2h/QWhut9WvAWUOqUQlCHBLCEYT8k22M/myt9SH79WHgbPv1LMAf0y9gtwnCMHw7fQS6A5JxIwh5ZsQzY7XWRillMj1OKXUDVngHrbVrqqu7sdK7GzXByHRVVVVxLHSMtv427rk6acGyguvKF27UBKIrE9yoySnZGvojSqlztNaH7NBMm91+EIitLlxntw1Da70cWG5vGrdUV3djpXc3aoKR6ZpaNpW1rWuZPnF6zt+bGz8vN2oC0ZUJbtRUW1vrqF+2oZvHgevs19cBj8W0X6uU8iil3gN0xYR4BCFK/YX1+Lv9kj8vCAUgraFXSj0IrAUuUkoFlFLXAz8CPqKU2gV82N4GWAm0ALuBu4Gv5EW1MKqJLHkgsXlBKAxpQzda608n2fWhBH0NcONIRQmlTXObNbVCvHlBKAwyM1YoKFIoRBAKjxh6oaBEvHlBEAqHGHqh4EhsXhAKixh6oeBIbF4QCosYekEQhBJHDL0gCEKJI4ZeEAShxBFDLwiCUOKIoRcEQShxxNALgiCUOGLoBUEQShwx9IIgCCWOGHpBEIQSRwy9UFC81d70nQRByCkjKiWolNoHdAMhYEBr/S6l1FTAB8wG9gFKa905MplCqSCGXhAKTy48+g9oredrrd9lb98CPKe1vgB4zt4WBBqbGqmrriu2DEEYc+QjdHMNcL/9+n7gE3m4hjAKaW5rpv7C+mLLEIQxx4hCN4ABVimlDPBru+D32TF1Yg8DZyc6UCl1A3ADgNbaNdXV3Vjp3Y2aIDNdDS80cPnsywvyPtz4eblRE4iuTHCjJqeM1NBfrrU+qJSaATyjlNoeu1NrbeybwDDsm8Jye9O4pbq6Gyu9u1ETZKZrzb41rLh6RUHehxs/LzdqAtGVCW7UVFtb66jfiEI3WuuD9u824FHg3cARpdQ5APbvtpFcQxAEQRgZWXv0SqkzgHFa62779ZXA94DHgeuAH9m/H8uFUGF04tvpY3fnbqkqJQhFZCQe/dnAGqXURuB14Cmt9dNHQ/AbAAAcbElEQVRYBv4jSqldwIftbWGMsrZ1LSBVpQShmGTt0WutW4BLErR3AB8aiSghOyb6fPTWFz6rZckTSwB4dvUcylta6FixAt9OH/UX1uPv9vOL9/+i4JoEQRhEZsaOcqYtWYK6fT7bP3sFZ9x7r6NjfDt9Obv+kieWsKdrD139XRzpOYLn5ElebX2VuzfdzZInlkjIRhBcwJg29I1NjVQ3NhZbRhwRPRN9yY1xY5PVJ2JkI5hJk5jo8+Hb6Ut4fMTAB7oDWWmLXDdyrnENDfi7/Zw/+XyeXfIsE+54gI3/+woALp1+KSAhG0FwA2Pa0De3NQOWUU1lWAtFrCGtXLs2aZ+jvUcHt//fR9Bf38Bb/ns13d/8Jnu3vETwv+/mjHvvjb6n2JtZdaM1OzXZDS5R+0SfL05bLN5qL3Mmz4luz/pyA5fVXsayK5ax4uoVKd6tIAiFYswa+samRhbMWEB3QwPlgew83FhebX01zhhGnhaSGchEeprbmulusDzg0PTpw4xu5FyLahcBsOLqFSy7Yll0f/9llzHze3dy4+lLMZMmUbl2LY1NjYxvtm5okVmp9RfWU3b0KEOJ9PXt9EWvHRvmifXO6y+sJ9zYOEyDIAjuY6QTpkYtzW3NUY9zoK6O3vp6fDt9/NNtrXDTTUmP8+308cWHd3Plxc3RtqO9R5k+cTqfGbeA8a++Spnfby3pFkN1Y2PUiMfS2NQYNaCx8ezuhoY4Qz/R56O5qtmRl9y1bFn0mgD9C+LPC9C3aFF0wDSW2L4ReuvrkQCMIIxexqSh9+30xRnVoZkqE30+ynfvZmDuXAD+dfZufvSsZSTXtq7li0wHoKu/K+64JR9roN8+vmFhA90LiRrIsqNHqW5s5DefnEv9hfU0NjXyo2eh2b5hJIplx94YygMBFlye2cBmd0MDDUD3wuH7euvrYcigbERzPdBtX1rWphGE0c+oN/STly6NpvQ5ZW3r2oQpf/UX1hO+rAbuuCPaVrl2Lcy2DPtEn4/ps6fT3dBAqqslSnHsW7SIKq3hk3Pj2p1mpQzU1dGwMLdGV4y4IIwNRr2hL29p4em+N7n/1x9m+Tu+R+ULL9Dd0BCdkdmwsGFYfvn0idNTnjO2b/nu3VFPN+KpZ0NvfT3lu3dHjetQj9/J8YIgCNlQEoOxxyoGeL36BP2XXQakTk2EzFL+YsMnIzW2iWL0giAI+WbUe/QDc+YQOronWrmou8Hy4GPDEuINC4Iwlhn1hr5r2TIWA4tj2sSwC4IgDFISoRtBEAQhOWLoBUEQSpy8hW6UUlcBtwNlwG+01imXK/7DH/7A1q1bmTRpErfckrqeeGdnJ7///e/p7e0lHA5z9dVXM2/evNyJFwRBKCHy4tErpcqAO4CPAvOATyulUlrihQsX8uUvf9nR+VetWsX8+fNZunQp1113HQ8//PCINQuCIJQq+fLo3w3sttesRyn1EHANsDXZAeeffz4dHR1xbe3t7TzyyCOcPHmS8ePHU19fz9lnW7XGT58+DUBvby+TJ0/Oz7sQBEEoAfJl6GcB/pjtABA3EV8pdQNwA4DWmpqaGowxcZXWly9fzhe/+EVmzpzJ7t270Vpz66238pnPfIYf//jHvPLKK/T19fHtb387Z9XZ3Vjp3Y2aQHRlghs1gejKBDdqckrR0iu11suB5famaW9v59ixYwwMDNDe3k5fXx+7du3i5z//efSYyL4XXniBd77znXzgAx9g7969/OpXv+Lmm29m3LiRR6LcWOndjZpAdGWCGzWB6MoEN2qqra111C9fhv4g4I3ZrrPbklJbW0t5eTkVFRXU1tbS09PDGWecwS9+MXxNmvXr13PrrbdSU1NDbW0tDz74INXV1TkL4Tj98AqJGzWB6MoEN2oC0ZUJbtTkhHylV74BXKCUOk8pNR74FPB4iv4ewPPVr371PL/fvwXwVFVVeY4fP75WKaUAj1JqnFJqPuAJBAJPf+UrX/m83T6vra3t0Je+9KVxkfOM5EcptT4X58nljxs1ia7Rr0l0jX5N9k9a8mLotdYDwFeBPwPbrCa9JdUxSqkHgbXARUqpgFLqeuDvgeuVUhuBLVgDugDfBL5ktz8IfE5rbfLxXgRBEEY7eYvRa61XAisz6P/pJLuuStB3K/DeLKUJgiCMKWRm7HCWp+9ScNyoCURXJrhRE4iuTHCjJkd4jJGIhyAIQinjltUr5W4jCIKQHWkHZN1i6GltbS22BMCdubJu1ATZ6TLhEJ5xZXlSZDFUl+k7jadyQl6vmY5S+g4LgRt1uVGT03RPidELheVAS05OY0Ih552PHc3JNQVhtCKGXigoZufmzIx0Mra/6bxvhxh6YWwjhl4oLIcPwvFjIz6NeX21877i0QtjnHwtU3yvUqpNKbU5H+cXRi+m7RB0HMnu2NM91u+BIOYvr2HCDp8MOt0VVxWEQpMvj/4+Ekx0EgTaDmGyDaW0HbJ+d3ZA7ykI7Hd2XMdRJI1YGMvkawmE1cDIn8+FksL091nedUdbdieIGPpjlodu9mx3dt2ek9DXm901BSED3OpQSIxeKBxH7ZCNA0NvjImGagDMwED0ScB02k8E+3Y5u25vD/ScykiqIGRFxBlxGUXLo09UeMQNuLG4gBs1Qea6+g7s5jhQ0X2cKUOOM/19UF6Bx64pEOo4SuhkF+Pr/gqA8PFjnAoPUF1Tw6nTPZwExvlbEl5/qK6OgSBnjq+gooifYal8h4XCjbrSaTLG0Pv6S1S97eICqnKGqwqPuAE3TopwoybIXFf4kFWSoH/7Zo5u2oDnnLroPnP4IJSX46mxSkWaTesxp04yruYca/tQAHP0CH3t7YTt2HwosI+jgQN4JlTFXWfatGlxZSlDJ09wvDWAZ9JZ2b3RHFAq32GhcKOudJpMx1HM7u30JOmTj8mCMmFKcB+nuq3ffb2E714Wv+/YUTgciG6awD5oj8nO6TmJOXXS2mfH6DEG9u8ZdpnQkSGzrCV0IxSC1gOYzo7k+3cULwkxX+mVidaWF8Y6tqEGwL8Xc3pwgNR0tmMOBzDhkNXu3xtv6E91D94oYtrNules3ydPRNuCu4aUPujrxYihH5OYHMzZcHyt1gMpU3kjcz9M3+lCSYqSl9BNirXlhbFMxFBHOLgfzn+L9brjKHR1QstOzMurLI/+zMFQi+k5Cae6MeFwvKFf8wzmkndjdm3B83efBWBg51Z46wJr/8AA9PdDT8xNRnCMMQZCA3jKK4otJSvM+lfwfOjqwlys9QAcT+7Rmx2bMMc7MM1r8Xzw44XRZDPqQzcmHC62BMEpp+KNrTm4b3Cjsx1z2I/Z0IR59Tk45B/i0Z+Ek91wohOC/YPtA0HCv/o+tA9m8gR3bR1Mc4ukVYpHnx2newezpUYR0TDfGy9H/xZMb0+qQ0Z+zUN+OHHcci6G7hsYgI42a+xpzTN51ZGIUW/oh3mJgmsxQ78r/77BfceOwqEAZuPrg/s72620yoEB63vu6U5sdEIhzPHBR+bQ4YPWTQGs+DxYE6wi18rxo3NJOxtdndnPeygS5nQv5kmf9Xezd+dgOGXHpvxd0xg4ctAaN+rqHN7hWBuEw5gnH7ImDRY43370G/oumZc1ahgSPjGBfYMbx9qhuytuQJZwmPAPboLN6yyPvL8fczDJbFh7EMyEQoRPHB9csfL0cI8+fPdPMYG9Vv+B4IjeEhCXzz+ajb6JfVKKcKITc8wdhn7okhexemMXyjNNL2G2bbCeCMNhOGAN2Jutf0l/jYMHMMEs/ia6uwb/xhLF6WMn+/WdHnRECsToN/THE9w9BXcy9I977w7Cy5dZA2bJFh47uB+za+vgk9v+3Yn7dXZYXlJ3l+VVRbzQyPo42zcS+tUPCP/6J7DxdcKPPWjt39Dk6JE+lQdm/vLa4MaW5rTnci3bNw0btDZdx12x+qc5fixuMpIJh2HrRuu1MfDmG9HX5sWV0OrH+K0lsU3LDkw4jLH7p7zOri2wdYMzTbE3xphML3P08PC+QydSdRxJGOLJF6Pe0JsTYuhHDUNDN6EQ5o2XCf/wW9Dfl/Qws2vrYMw12WzYgeBgDB8wHW2EH/7tYOjmWDtsfB2zbo21vfF1zIlOzNYN1uN97PUOBazB38i2MbCxKbG2nlOYbbbBCYcIv/R00vfhdsyBPWA/6UQ5kT50Y7q78qjKpmV73DgMHW2Y/fbfwrGjmM3rrde7tkBgH5gw5vWXLX1/ec1yJNoOpV8i++D+wb+RFJjTPXHpkib2SXTTOqtt64bBp5C9Q/5u249gml9Ne51cMeoNPSeOj+rH5bGCCYeHDcZGOZZmYsyBPYOPw60HkvfrbIeu49b1nnkMs+pRzOYkHrYJY95Yg9m2EdMyuGaOCYUI//rH8fn5J08QfvaJwT52yMn0nbbisof81vvz74OYgWBzohPTPnxMIeLdpTI6JhjMLoQwAsyBPRj/EEPf1ZlymWfTdxrz0G/yrAzMnh2Y2BvO4QAcCtj7tg9+Jy/+abDP5nV234OY9a+ACcfFz004TPiJh+LCdyawD7OxKeFnb4yxZnAD7N9jDb5GOHJwsN+b6zBHDxP+5fcwD/0GE+zHbHgt/lxHWq0njwIx+g19KAQxOdSCSznda/2jZUMoZHlpYMVck9HZMfiEZ98YzMurknY3zzxmeVYtMR793p2WV3dgj/WPHey3QgY7NmGOHsac6ib8uzsxh/yYR+7DHApYTyOd7Zidm61xCDt3O/y7uwg/uHz4ddetsQz5Gy8PPg0c74g3LpvegNispEJwoAX8QyqAneiMhm6MMdGB7Ggoa+sGzLqXM85XN7u3Jd839LMAzJ5tlhcfDlmhmFb/oKHds936znp7MLFPXjGhEfPnR60XMfFz89oLmMf/EM2CMcZYKb+9PbDjTcyRVsIvr4o6kqFDAcwfH7D67t0ZP8Hv8KChp6+X8H/9GEIDmNVPY154anCsKNJ/zTNWZlmBGP2GHqKP64KLKUB2lGk/YhU2iSXRAGOEiIe4cwvh1U9bhiwyYLd/D7TsgJ1bBj3wppdg9zbYs53wXT/CNL0YNYxmSzPm2cetYw/usyZ9bW6O+2eOLtJ2oAVa98OebYTv+6VlUP77TsyfHo72Db/2IibBrN+hhO/+KaE7/iPe0GCHu+ynHycZHubUSeum599reZu2oTVdx+F4h/X0sfF1zOurLWO/0wpbmI1NVjbJ2ufTXiNO91M+62aaaLbonu0Etw3G082pk9b30dFmeecvrrQ+18MHrSeKTeug77T1+fcn+b4j4aXYG1LkPTxyP+Hf3QkbX49mZ5m/NGGe8mEe+BXmSR8A/ZvWY55/0kqR3LsLc/ggJhwm9G9fsY6NxR4AJhTCPHL/cD3tR+Bkd26SARzgmuLgI+J4J9SdV2wVwhBMz0k8VZOsjQJMWDIPDfeeHdHXi/nvO6F8vBWzB8z+3fDaizDpTMAOxTS9NJimaRvwqDf4u/+KPrGYDU3WP3Kw3zKefX30bXid8F0/ZtwP/ssywPv3YHZvh2NHCX/v69DfZ90sLl0ElZVWnHfSmanf76mTmDfWgAkT7ulm3Lf+A4/HQ/jpFZgV98PU6db19D3wmS/j8XgSn+dEJ+E7f2htHGgh/K//F89lH4LPfc1yosJhwr/6AbQfhpqZML4SjrXjuegdUUNt1r6AuWpJ0mvEXe90D2x/E7PyEcy2DYy75Sd4zvFijh/Dc9ZUzP499B3yY3osL9hs3wgDQSt0s28Xpnkt1Jxtecz3/RLswU/zzP+kv/bxDszqpxl3xVXW5w/W9//S05iY8RWzbs3gQP6TD2He8g76N60HY6wB/bIyqKiwxg5i4/MJL5riKbTrOEybnlb3SCkJQ2+6Okn/5yUUGvP8k7BYWStSjoIqT+bhewfHEdoOYbpPwOy5eKonW22HA3FLLQCDj+Qx/8yxBgNj4HCA3heftEJA+3ZBW6vlTUZSRSNx31CI8L2/sAaWBwaGefRmIAj7dsOcC6HzmJWBFLnuzi2wpRnzlosHwxTHjhK+64ewaR2e+QvhbZfGn+9UN2bKFMwj91nhj9h9rz4H8+YPxrQjg51Hj2D6+/BMmWYNwkbGIA754cAezF+dD50deKbW2G/fDDP+Zs0z1vtbby9fse4VuOgdhO/6D8bd/BPL0HcdI7z+VTut1p73YBt6+k5HP7u4gVMn2VPbNsKb6zA1M+Pi6sOIdUyMIXz/r+iPPJHF1DYwa55Ne82UHO8QQ++YHIRujDHJH/ucHJ+HlelGG0P/qc2b6/C8412YmpmE9b1FVOaQoUa89xS07MCcXZu8jwNMYC/9f7Ee7c0zj1mP8+tfTezpxWa9HNxv3WzOOAOCQcwrz2KeeBDP2xZg/HvxnHdB3KHh392F5xN/H6/RzgAJ/+kRxs2bD9s2QuUEPOe/BfPin+jcvhGzPfFEIvPg8uEzikMDsHMzpm42nn3xqa7hZQ1wTh1MnkLZV/8VEwwS/vHNMG0G4770LSvk9cz/DAtzmOefxDylLQ+96SU4sIdQos+5qzNlbN8R9rXDv78rs+PaWkkUADOvvTgyPYkmV+WB0jD0QzMFsuFwwIr1feZLKbuFX30ez6IPDPdSnvDB337a0aNrKWLCYesx+rp/wlNWZnmf/r2Yda9YA2kJcotHBX2nrZj6CDAvr4qOUUQWYcNJbDY0QPjfvwZnTrYHRA30nLKMIURj8FE62jD3/2fic+3YBM2vEv7DryHYz7if3IvZtI5gqipdqcZVDgWGG92+XsvjLi+3Bkaf8llPHft3E/7JLda+ROMFMUbdvLgy+c3UmNz8r0PuCoSERpYLX6hoRN4GY5VSVymldiildiulbsnXdQDM+lcSTlJI2j8cHjZAZTatxzz7RNxovznRaXlesdv3/XJYJoc5dRLz9CPWwN3p4SXrzP7dI0oBNeFwXNk8kyLnPEL4z3+Mny24bk10MDC8+s/xs1IjfXZsInzvLxK+h6TaInHMppcwa5/H/OlhayJI6wErrvr0Cti11fH5SpJYY5pp5lHXMcu49Zx0tl5Pikk44bt/CieOQ28PZtX/WIPN2RIasAajk2gw9//nYAgJrGwmJ9P+x1oGXYFm9pd997vfzflJlVJlwNPA3wA/BH758MMPr/7kJz+ZLCH3u93d2WVlmF1bYNtGTMdRPJcuxFM2PHxigv1x7eYpH2blI1aqngnDmZMxjz8I/r2cfvUFwhMmQjBI+J7bMK88i2fWuXCsHbN7qzUDb98uPO//qOWheMZhXnkWtvwFs3m9la51YA+eCROtuGXfacL/+X2YPAWPPWBs+vusMFHHEcsInnEGZuMbhO/8IZ4ZM2FilTU7b+YsPB4PZZvXcfr+O/C87yrYs4Pwnf9hPVWUl2O6T2B+f6eVZ+xvseK3m5sxv70dqiZBeTmcOZnwnT+yBvxOdFqZBJvXw4SJMKUGT2Ul5nDAGmTauRlOncRzyV9bWjs7rFj7rHOtvPQ9O+C8C/CMK2N8Zzs9/3K99ST0nJ1nvmMT5uU/W6lvLi2rNmaJNbS7tjozvKnoTXHjKWDq4GjGc3atNX6SJdXV1QD/nvY6+VhcRym1CPiu1vpv7O1vA2itf5jkENPa2ppkV2rCT2nM//zO2phxDpzjxTNtBp4Fl1l5sh1tsHcnnk9/Gc9f/y9rWduHlsfnY0+eankS6R7DqiYNDtKcPSv1YM5QzqiG6TPxXPLXmA2vx0/ln1gFwaD1OF85wdLT1mr9nj2XsrZDhA754aJ3WMed7oXpM61lfA8F4geOZp1redOx36v3vOSPvGVlMGmyNc4Re8x5F1qZEJ3tVlx3YtXgYNf0mXgu/ms8b75OeBSubCgIruHS91D2lVuzPtyuMJU2+pOvGP0sIPaWHgDiblu5qhkbvOLDDMzyDt/R2w2XvMveWGz92rIOpkyBf7w56fnGjRtHON8zbevOzah7QTSl4oOLEzaPm3dxcXUloeifVwLcqAlEVybkQ9O46TOpLEBt3NFfM7Z6KsxflCtZo7JWZbEQXc5xoyYQXZmQL03dIzin05qx+TL0B4FYN7vObkuKU8GFwE1aIrhRE4iuTHCjJhBdmeBGTU7IV9bNG8AFSqnzlFLjgU8Bj6fo73HLj1JqfbE1jAZNomv0axJdo1+T/ZOWvBh6rfUA8FXgz8A2q0lvSX2UIAiCkA/yFqPXWq8ECrcOpyAIgpCQ0li9MrdkuTJWXnGjJhBdmeBGTSC6MsGNmhyRlzx6QRAEwT2IRy8IglDilMaiZilQSnmBB4CzsRYWX661vl0pNRXwAbOBfYDSWncqpd4C/BZYADRorX8ac657gY8DbVrrtxdbU7LzuEDXBGA1UIn1N/aI1vo7xdYVc74yYB1wUGv98WJrUkrtA7qBEDCgtX4XWZJjXWcBvwHebp/rC1rrtcXUpZS6yO4fYQ7w/7TWvyiWJvtcNwFftM+zCfi81vp0ppryxVjw6AeAb2qt5wHvAW5USs0DbgGe01pfADxnbwMcA74G/DTBue4DrnKRpmTnKbauPuCDWutLgPnAVUqp97hAV4SvY2WDjYRca/qA1nr+SIx8HnTdDjyttX4LcAkj+8xyoktrvcP+nOYD7wR6gEfJjpxoUkrNstvfZTuAZVgp5a6h5A291vqQ1rrZft2N9cc6C7gGuN/udj/wCbtPm9b6DWDYOrJa69VYX7YrNKU4T7F1Ga11ZAGeCvsn68GgXH6HSqk64GNYnmrW5FJTLsmVLqXUZOAK4B67X7/W+nixdQ3hQ8AerfV+F2gqByYqpcqBKiC7xbvyRMmHbmJRSs0GLgWagLO11pHlFQ9jPb6NWk1DzlN0XXZ4ZD0wF7hDa+0KXcAvgH8BqnOhJ0eaDLBKKWWAX9vLgxRb13nAUeC3SqlLsL7Lr2utHayVnFddsXwKeHCkekaqSWt9UCn1U+AA0Aus0lonr0pfBEreo4+glJoErAC+obWOW/Raa20YgcdZbE2pzlMsXVrrkP14XQe8WymV9ZhGrnQppSLjK+tHqiVXmmwu11ovAD6KFT64wgW6yrFi0XdprS8FTjEYwiimrsh5xgN/Czycrm++NSmlpmA9BZwH1AJnKKX+YaS6csmYMPRKqQqsL/L3Wus/2s1HlFLn2PvPAdpGo6Yk5ym6rgj24/4LjHBsI0e63gv8rT34+RDwQaXU74qsCa31Qft3G1a8+d3ZasqhrgAQiHkSewTL8BdbV4SPAs1a6xGtk50jTR8G9mqtj2qtg8AfgctGoivXlLyhV0p5sOKM27TWP4/Z9Thwnf36OuCx0aYpxXmKrWu6nbGBUmoi8BEgRc26wujSWn9ba12ntZ6N9dj/vNY6K88rh5/VGUqp6shr4EpgczaacqlLa30Y8NtZLmDFw7MuFZaH/8NPM8KwTQ41HQDeo5Sqss/5IUY+2J9TSn7ClFLqcuBlrJSnyGLSt2LF4jTwV8B+rBSqY0qpmVipd2fa/U8C87TWJ5RSDwLvB2qAI8B3tNb3FEsTcHGi82hr+YmMyaGu2ViDWGVYzoTWWn8vG0251BX7WK6Uej/wLZ19emWuPqsaBrNGyoE/aK0bs9GUS1323/t8rEHr8UALVspgVtWsc6zrDCzjOkdr3ZWNnjxo+negHiuT5y/AF7XW6Wt+FoiSN/SCIAhjnZIP3QiCIIx1xNALgiCUOGLoBUEQShwx9IIgCCWOGHpBEIQSRwy9IAhCiSOGXhAEocQRQy8IglDi/H8ILUTl17EUTAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create and visualize candlestick charts\n",
    "ax1 = plt.subplot2grid((6,1), (0,0), rowspan=5, colspan=1)\n",
    "ax2 = plt.subplot2grid((6,1), (5,0), rowspan=1, colspan=1, sharex=ax1)\n",
    "ax1.xaxis_date()\n",
    "\n",
    "candlestick_ohlc(ax1, df_ohlc.values, width=2, colorup='g')\n",
    "ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Candlestick graphs help condense __price direction__ and __open-high-low-close__ data all in just one graph. In addition, we have plotted the __volume__ in the lower subplot to condense even more information in just one plot."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"SP\"></a> Automating Getting the S&P 500 List\n",
    "\n",
    "The [S&P 500](https://en.wikipedia.org/wiki/List_of_S%26P_500_companies) is a list of (generally) the top 500 companies by _market capitalization_. The market capitilization describes the value of a company, and is calculated as the product of the number of its outstanding shares and the price of each share. The S&P can therefore be thought of as the __top 500 most valuable companies__. The performance of the S&P can be used as an indicator for roughly how well the economy is doing. Investors commonly refer to the S&P as a benchmark of the overall market, to which all other investments are compared. \n",
    "\n",
    "Here we will build create a ___web scraper___ that will collect the S&P data from the internet. To do this, we use a web scraping library called [Beautiful Soup](https://www.crummy.com/software/BeautifulSoup/).\n",
    "\n",
    "By first opening our [website](https://en.wikipedia.org/wiki/List_of_S%26P_500_companies) that contains the data we want, we take note of how the data is presented. We observe that it is displayed in a _table_, and the first row second column has a security, '_3M Company_'. What we want is to collect the data that is displayed in this table. To do this, we can extract it from the websites source code. To view the source code we first go to the website, and then press ```CTRL+U``` on the keyboard. By then pressing ```CTRL+F```, we search for the occurence of the first elements in the table (e.g. we search for the term '3M Company'), to get an idea of where the table data begins. By then searching for the term '_table_', we observe an occurence of it only a couple of lines above it. A couple of lines below that, we observe the first table data tags, ```<td>```, this denotes the beginning of the table data that we want to extract.\n",
    "\n",
    "Let's collect that data!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import bs4 as bs  # Beautiful Soup (web scraping library)\n",
    "import pickle  # Allows us to save S&P list\n",
    "import requests"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The [pickle](https://docs.python.org/2/library/pickle.html) module serializes any Python objects, and allows us to __save any object__ as a variable. Here we will use it to save the S&P 500 list, so that we won't have to always go back to wikipedia to collect that data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saves the S&P 500 ticker data\n",
    "def save_sp500_tickers():\n",
    "    resp = requests.get('https://en.wikipedia.org/wiki/List_of_S%26P_500_companies')\n",
    "    soup = bs.BeautifulSoup(resp.text, 'lxml')\n",
    "    table = soup.find('table', {'class':'wikitable sortable'})\n",
    "    tickers = []\n",
    "    \n",
    "    # Iterate through the table\n",
    "    for row in table.findAll('tr')[1:]:  # [1:] Skips the table header\n",
    "        ticker = row.findAll('td')[0].text  # Collects the ticker data\n",
    "        tickers.append(ticker)\n",
    "    \n",
    "    # Save the collected data\n",
    "    with open('sp500tickers.pickle', 'wb') as f:\n",
    "        pickle.dump(tickers, f)\n",
    "        \n",
    "    return tickers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['MMM', 'ABT', 'ABBV', 'ABMD', 'ACN']"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "save_sp500_tickers()[:5]  # Print just the first 5 tickers in the table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparing these with the first 5 tickers in the table of our [wikipedia page](https://en.wikipedia.org/wiki/List_of_S%26P_500_companies), we observe they are identical. Having sucesfully scraped the ticker data for our S&P 500 list, we will now collect the rest of the relevant data.\n",
    "\n",
    "It appears that the morningstar API gets stuck loading a few of the tickers. I have added a try and except code block, so that if it takes too long, interrupt the kernel so that it will continue loading the others."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data_from_web(reload_sp500=False):\n",
    "    if reload_sp500:\n",
    "        tickers = save_sp500_tickers()\n",
    "    else:\n",
    "        with open(\"sp500tickers.pickle\", \"rb\") as f:\n",
    "            tickers = pickle.load(f)\n",
    "    if not os.path.exists('stock_dfs'):\n",
    "        os.makedirs('stock_dfs')\n",
    "\n",
    "    start = dt.datetime(2010, 1, 1)\n",
    "    end = dt.datetime.now()\n",
    "            \n",
    "    # Check if file already exists as a locally saved file    \n",
    "    for ticker in tickers:\n",
    "        try:\n",
    "            # Collect and save data locally if it doesn't exist yet\n",
    "            if not os.path.exists('stock_dfs/{}.csv'.format(ticker)):\n",
    "                df = web.DataReader(ticker, 'morningstar', start, end)\n",
    "                df.reset_index(inplace= True)\n",
    "                df.set_index('Date', inplace= True)\n",
    "                df = df.drop('Symbol', axis=1)\n",
    "                df.to_csv('stock_dfs/{}.csv'.format(ticker))\n",
    "            else:\n",
    "                print('Already have {}'.format(ticker))  # Already exists\n",
    "        except:\n",
    "            print('Cannot obtain data for ' + ticker)  # If it takes too long"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get_data_from_web()  # Collect and store all ticker data locally"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Combining S&P 500 into one DataFrame\n",
    "\n",
    "Having collected all our necessary csv files from the web, we will combine them neatly into one DataFrame object. We will combine them such that we record only the _Close_ data for each ticker.\n",
    "\n",
    "Some tickers failed to download, so we will ensure to skip those when combining the dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "missing_tickers = ['ANDV', 'BKNG', 'BHF', 'CBRE', 'DWDP', 'DXC', 'EVRG', 'JEF', 'TPR', 'UAA', 'WELL']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compile_data():\n",
    "    with open('sp500tickers.pickle', 'rb') as f:\n",
    "        tickers = pickle.load(f)\n",
    "    \n",
    "    main_df = pd.DataFrame()\n",
    "    \n",
    "    for count,ticker in enumerate(tickers):\n",
    "        if ticker in missing_tickers:\n",
    "            pass\n",
    "        else:\n",
    "            df = pd.read_csv('stock_dfs/{}.csv'.format(ticker))\n",
    "            df.set_index('Date', inplace=True)\n",
    "\n",
    "            df.rename(columns = {'Close': ticker}, inplace=True)\n",
    "            df.drop(['Open','High','Low','Volume'], 1, inplace=True)\n",
    "\n",
    "            # Join the DataFrames together\n",
    "            if main_df.empty:\n",
    "                main_df = df\n",
    "            else:\n",
    "                main_df = main_df.join(df, how='outer')\n",
    "\n",
    "            if count % 50 == 0:  # Track progress\n",
    "                print(count)\n",
    "            \n",
    "    print(main_df.head())\n",
    "    main_df.to_csv('sp500_joined_closes.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "50\n",
      "100\n",
      "150\n",
      "200\n",
      "250\n",
      "300\n",
      "350\n",
      "400\n",
      "450\n",
      "500\n",
      "              MMM      ABT  ABBV  ABMD    ACN   ATVI   ADBE   AMD    AAP  \\\n",
      "Date                                                                       \n",
      "2010-01-01  82.67  25.8328   NaN  8.73  41.50  11.11  36.78  9.68  40.48   \n",
      "2010-01-04  83.02  26.0577   NaN  8.74  42.07  11.30  37.09  9.70  40.38   \n",
      "2010-01-05  82.50  25.8471   NaN  8.53  42.33  11.32  37.70  9.71  40.14   \n",
      "2010-01-06  83.67  25.9907   NaN  8.40  42.78  11.26  37.62  9.57  40.49   \n",
      "2010-01-07  83.73  26.2060   NaN  8.40  42.74  10.99  36.89  9.47  40.48   \n",
      "\n",
      "              AES ...    WYNN    XEL      XRX   XLNX     XL  XYL      YUM  \\\n",
      "Date              ...                                                       \n",
      "2010-01-01  13.31 ...   58.23  21.22  22.2887  25.06  18.33  NaN  25.0967   \n",
      "2010-01-04  13.67 ...   63.96  21.08  22.7365  25.38  18.72  NaN  25.1828   \n",
      "2010-01-05  13.53 ...   67.85  20.83  22.7629  25.06  18.59  NaN  25.0967   \n",
      "2010-01-06  13.39 ...   66.96  20.87  22.5521  24.89  18.40  NaN  24.9173   \n",
      "2010-01-07  13.40 ...   68.39  20.78  22.6575  24.64  18.40  NaN  24.9101   \n",
      "\n",
      "              ZBH   ZION  ZTS  \n",
      "Date                           \n",
      "2010-01-01  59.11  12.83  NaN  \n",
      "2010-01-04  60.02  13.33  NaN  \n",
      "2010-01-05  61.92  13.80  NaN  \n",
      "2010-01-06  61.90  15.00  NaN  \n",
      "2010-01-07  63.32  16.68  NaN  \n",
      "\n",
      "[5 rows x 494 columns]\n"
     ]
    }
   ],
   "source": [
    "compile_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that not all of the tickers were trading during early 2010, therefore several of the tickers have missing data (NaN) when observing the head of the dataframe."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### <a class=\"anchor\" id=\"corr\"></a> S&P 500 Company Correlation Table\n",
    "\n",
    "In this section we will briefly investigate any relationships/ patterns between companies that may be present within this data.\n",
    "\n",
    "The correlation between companies can be of interest to investors. If you have two companies that are highly correlated, and once one of them starts to deviate, you can invest in one and short the other one and then eventually they come back together. To give another example, if a pair of companies are negatively correlated and they both go in one direction, then you can again invest in one and short the other. \n",
    "\n",
    "Investors ideally want to be properly diversified. This can be achieved by investing in non-correlated stocks, and helps to minimize the risk of having all of your investments performing poorly at the same time.\n",
    "\n",
    "Here we will visualize correlations between the S&P 500 companies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_data():\n",
    "    df = pd.read_csv('sp500_joined_closes.csv')\n",
    "    df_corr = df.corr()  # Obtain correlation data\n",
    "    data = df_corr.values\n",
    "    \n",
    "    # Visualization\n",
    "    fig = plt.figure()\n",
    "    ax = fig.add_subplot(1,1,1)\n",
    "    heatmap = ax.pcolor(data, cmap=plt.cm.RdYlGn)\n",
    "    fig.colorbar(heatmap)\n",
    "    ax.set_xticks(np.arange(data.shape[0]) + 0.5, minor=False)\n",
    "    ax.set_yticks(np.arange(data.shape[1]) + 0.5, minor=False)\n",
    "    ax.invert_yaxis()\n",
    "    ax.xaxis.tick_top()\n",
    "    \n",
    "    column_labels = df_corr.columns\n",
    "    row_labels = df_corr.index\n",
    "    \n",
    "    ax.set_xticklabels(column_labels)\n",
    "    ax.set_yticklabels(row_labels)\n",
    "    plt.xticks(rotation=90)\n",
    "    heatmap.set_clim(-1,1)\n",
    "    plt.tight_layout()\n",
    "    plt.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <a class=\"anchor\" id=\"form_strat\"></a> Preprocessing Data for Machine Learning\n",
    "\n",
    "In the previous section we realised some clear relationships between companies. The question we ask in this section is whether we can get a machine to recognise those patterns. \n",
    "\n",
    "A hypothesis we can test is that the share price of groups of companies are likely to move together, and that we can therefore use the price movement of other companies as an indicator for a prediction in which direction a single company's share price will move in the future.\n",
    "\n",
    "We will preprocess our data such that we use a normalized _percentage change_ in the stock price as a feature to make a prediction. <a class=\"anchor\" id=\"strat\"></a>  \n",
    "\n",
    "Our target variables (labels) are three classes of options: buy, sell, or hold. Our aim is to use our training data (the percentage change in the stock price) to train a machine to make optimal decisions in response to the question: 'did the price based on these features within the next 7 days of trading go up more than 2%?' If the answer is yes, then buy. If instead it fell by 2%, then sell. If it did neither of those things, then we hold. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_data_for_labels(ticker):\n",
    "    hm_days = 7  # How many days for price to go up/down X%\n",
    "    df = pd.read_csv('sp500_joined_closes.csv', index_col=0)\n",
    "    tickers = df.columns.values.tolist()\n",
    "    df.fillna(0, inplace=True)\n",
    "    \n",
    "    # Obtain percentage change within given number of days \n",
    "    for i in range(1, hm_days+1): \n",
    "        df['{}_{}d'.format(ticker, i)] = (df[ticker].shift(-i) - df[ticker]) / df[ticker]\n",
    "        \n",
    "    df.fillna(0, inplace=True)\n",
    "    return tickers, df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating Machine Learning Target Function\n",
    "\n",
    "Here we create a new column for our price % change dataframe, which encodes the target variable whether the company is a: buy/ sell/ hold."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def buy_sell_hold(*args):\n",
    "    cols = [c for c in args]\n",
    "    requirement = 0.02  # 2% change in share price\n",
    "    \n",
    "    # If price goes up 2% ...\n",
    "    for col in cols:\n",
    "        if col > requirement:\n",
    "            return 1  # Buy\n",
    "        if col < -requirement:\n",
    "            return -1  # Sell\n",
    "    return 0  # Hold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating Labels for Machine Learning\n",
    "\n",
    "Here we map the target function created previously to our dataframe of share price % changes to create a new column of decisions of whether to buy/ sell/ hold a companies shares."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "def extract_featuresets(ticker):\n",
    "    tickers, df = process_data_for_labels(ticker)\n",
    "    \n",
    "    # Create new column of classes: buy/ sell/ hold \n",
    "    df['{}_target'.format(ticker)] = list(map(buy_sell_hold,\n",
    "                                              df['{}_1d'.format(ticker)],\n",
    "                                              df['{}_2d'.format(ticker)],\n",
    "                                              df['{}_3d'.format(ticker)],\n",
    "                                              df['{}_4d'.format(ticker)],\n",
    "                                              df['{}_5d'.format(ticker)],\n",
    "                                              df['{}_6d'.format(ticker)],\n",
    "                                              df['{}_7d'.format(ticker)],\n",
    "                                             ))\n",
    "    \n",
    "    vals = df['{}_target'.format(ticker)].values.tolist()\n",
    "    str_vals = [str(i) for i in vals]\n",
    "    print('Data spread:', Counter(str_vals))\n",
    "    \n",
    "    df.fillna(0, inplace=True)\n",
    "    df = df.replace([np.inf, -np.inf], np.nan)\n",
    "    df.dropna(inplace=True)\n",
    "    \n",
    "    df_vals = df[[ticker for ticker in tickers]].pct_change()\n",
    "    df_vals = df_vals.replace([np.inf, -np.inf], 0)\n",
    "    df_vals.fillna(0, inplace=True)\n",
    "    \n",
    "    X = df_vals.values  # Features\n",
    "    y = df['{}_target'.format(ticker)].values  # Target labels\n",
    "    \n",
    "    return X, y, df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extract_featuresets('MMM')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <a class=\"anchor\" id=\"an_strat\"></a> Machine Learning\n",
    "\n",
    "Having preprocessed our dataset, we are now ready for machine learning.\n",
    "\n",
    "Recall that our target variable, y, takes values: \n",
    "\n",
    "$$y=\\{-1,0,1\\}=\\{\\text{sell}, \\text{hold}, \\text{buy}\\}$$\n",
    "\n",
    "We want to use our feature data, $X$ (the percentage change in each of the S&P 500 company share prices), to predict whether we should make the decsion to buy, sell, or hold a given share for a given period of time - based on the trading strategy of whether the price has moved up/ down 2% as discussed [previously](#strat)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n",
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "from sklearn import svm, cross_validation, neighbors\n",
    "from sklearn.ensemble import VotingClassifier, RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def do_ml(ticker):\n",
    "    X, y, df = extract_featuresets(ticker)  # Extract features\n",
    "    \n",
    "    X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.25)\n",
    "    \n",
    "    clf = neighbors.KNeighborsClassifier()\n",
    "    clf.fit(X_train, y_train)  \n",
    "    confidence = clf.score(X_test, y_test)\n",
    "    print('Accuracy', confidence)\n",
    "    predictions = clf.predict(X_test)\n",
    "    print('Predicted spread:', Counter(predictions))\n",
    "    \n",
    "    return confidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data spread: Counter({'1': 999, '-1': 925, '0': 307})\n",
      "Accuracy 0.471326164874552\n",
      "Predicted spread: Counter({1: 347, -1: 198, 0: 13})\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.471326164874552"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "do_ml('BAC')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we see that the k-Nearest neighbours algorithm can accurately predict a buy/sell/hold opportunity with a ~40% accuracy. \n",
    "\n",
    "We will improve upon this model by using a method known as ___ensembling___. This involves combining several models together and mixing their predictions together to give a (typically better) overall weighted prediction. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def do_ml(ticker):\n",
    "    X, y, df = extract_featuresets(ticker)  # Extract features\n",
    "    \n",
    "    X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.25)\n",
    "    \n",
    "    # Instantiate ensemble model\n",
    "    clf = VotingClassifier([('lsvc', svm.LinearSVC()),\n",
    "                           ('knn', neighbors.KNeighborsClassifier()),\n",
    "                           ('rfor', RandomForestClassifier())])\n",
    "    \n",
    "    clf.fit(X_train, y_train)  \n",
    "    confidence = clf.score(X_test, y_test)\n",
    "    print('Accuracy', confidence)\n",
    "    predictions = clf.predict(X_test)\n",
    "    print('Predicted spread:', Counter(predictions))\n",
    "    \n",
    "    return confidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data spread: Counter({'1': 999, '-1': 925, '0': 307})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy 0.4265232974910394\n",
      "Predicted spread: Counter({-1: 326, 1: 232})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.4265232974910394"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "do_ml('BAC')  # Bank of America = BAC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we observe that the accuracy of the overall model has increased, but not significantly. The model could be further improved upon by tuning the hyperparameters in each of the models appropriately.\n",
    "\n",
    "Interestingly, in both models we observe far fewer predictions for a _hold_. This is most signifcant in the enesmebled model, where not a single prediction was made for a hold.\n",
    "\n",
    "It should be noted that the dataset we have used was quite highly unbalanced, with far fewer holds than buy/sells. This could have lead to a skew in our predictions, which could have indeed been a cause for the significantly fewer _hold_ predictions.\n",
    "\n",
    "To help balance our dataset, we make changes to the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def buy_sell_hold(*args):\n",
    "    cols = [c for c in args]\n",
    "    requirement = 0.03  # 3.0% change in share price\n",
    "    \n",
    "    # If price goes up 3% ...\n",
    "    for col in cols:\n",
    "        if col > requirement:\n",
    "            return 1  # Buy\n",
    "        if col < -requirement:\n",
    "            return -1  # Sell\n",
    "    return 0  # Hold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data spread: Counter({'1': 795, '0': 735, '-1': 701})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy 0.3727598566308244\n",
      "Predicted spread: Counter({-1: 268, 1: 173, 0: 117})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/anthony/.local/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.3727598566308244"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "do_ml('BAC')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately, after using a balanced dataset we observe an accuracy that is not much better than just random predictions. Since there are only three choices of class to predict, if we were to predict (for example) _buy_ for every single situation we would have arrived at roughly the same accuracy as our model. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <a class=\"anchor\" id=\"quant\"></a> Quantopian \n",
    "## Algorithmic Trading with Python\n",
    "\n",
    "[Quantopian](https://www.quantopian.com/) is a website and environment that __simulates trading__, providing stock pricing accurate to the minute. It is a useful resource for [backtesting](https://www.investopedia.com/terms/b/backtesting.asp), a method used to assess the viability of a trading strategy by testing how it well it would perform with historical data. Quantopian has changed a lot over time, and has now become more of a __research platform__ that you just so happen to backtest on. \n",
    "\n",
    "When writing algorithms in Quantopian, the variable ```context``` is a python dictionary that contains information about your portfolio. It can tell you about the portfolio performance, your leverage, your account, etc. The Quantopian variable ```data``` is your universe of information, containing numerous data such as stock prices. \n",
    "\n",
    "Quantopian has excellent [documentation](https://www.quantopian.com/help) which can be viewed at [Quantopain Help](https://www.quantopian.com/help).\n",
    "\n",
    "The following [notebook](https://www.quantopian.com/posts/how-to-get-an-allocation-writing-an-algorithm-for-the-quantopian-investment-management-team) and its associated [YouTube webinar](https://www.youtube.com/watch?v=bvUdDFpsZ3o&feature=youtu.be) are excellent introductions to Quantopian and several of its main tools, including:\n",
    "1. [Blaze](http://blaze.pydata.org/) - study and interact with partner data.\n",
    "2. [Pipeline](https://www.quantopian.com/tutorials/pipeline) - dynamically select stocks to trade each day.\n",
    "3. [Alphalens](https://www.quantopian.com/posts/alphalens-a-new-tool-for-analyzing-alpha-factors) - analyze the predictive ability of an alpha factor.\n",
    "4. [The Interactive Development Environment](https://www.quantopian.com/algorithms) (IDE) - backtest the strategy.\n",
    "5. [Pyfolio](https://www.quantopian.com/posts/how-to-get-an-allocation-writing-an-algorithm-for-the-quantopian-investment-management-team) - understand the performance and risk analysis of the backtest.\n",
    "\n",
    "NOTE: Much of the code in this Quantopian section will not run correctly, because they depend on packages which (as far as I know) can only be accessed through Quantopian's website (such as ```pipeline```). It is therefore recommended to follow the notes in this notebook alongside the outputs of the code in [my Quantopian notebook](https://www.quantopian.com/research/notebooks/Q%20Research%20Tutorials.ipynb).\n",
    "\n",
    "## <a class=\"anchor\" id=\"orders\"></a> Orders and Leverage\n",
    "\n",
    "Quantopian offers several [order methods](https://www.quantopian.com/help#api-order-methods), which can be used to place buy/sell orders in a simulated environment on financial markets using the algorithm you are building. The method [```order_target_percent()```](https://www.quantopian.com/help#api-order-target-percent) is considered to be the most useful, and easiest to use to keep your portfolio balanced with the shares you want to own. By specifying a positve value for its ```style``` parameter, you can place long orders on the market to fill your portfolio to reach the specified balance as determined by  ```style```. By having a _negative_ ```style``` value, you can short the companies shares to remove those shares from your portfolio.\n",
    "\n",
    "[Shorting](https://www.investopedia.com/university/shortselling/shortselling1.asp) is essentially a way of __betting against a company__. This is achieved by _borrowing_ some shares from someone, selling them at the current price, and then finally repaying those shares back to them in the future in hopes that those shares have decreased in value. If done succesfully, this allows you to make a profit by selling borrowed shares at a high price and repaying them back at a much lower price in the future. \n",
    "\n",
    "The __danger with shorting__ is that there is __infinite downside__, with only a __finite upside__. As opposed to investing and going long, where the share price can go up infinitely (and the downside is finite), the most you can earn by shorting is the price at which you borrowed those shares (in the absolute best case scenario that the shares go to zero, and you repay them back at that price). In addition to this, shorting is dangerous because you are __borrowing money__.\n",
    "\n",
    "In Quantopian, the functions ```initialize()``` and ```handle_data()``` run automatically without ever needing to be called. ```initialize()``` runs just once, when the program is executed. However, the function ```handle_data()``` runs once __every minute__. This can lead to tricky  bugs, and needs to be accounted for. For example, if you set a buy order for a million dollars worth of shares, that can take some time to fill. If an event occurs that triggers a sell order for a million dollars worth of shares before those shares have been filled, the next minute when ```handle_data()``` has been called, then this would result in unwanted short selling, where you are borrowing and selling shares that you do not own. The effects of this can get amplified over time, where your order book continues to grow exponentially due to ```handle_data()``` being called every minute. To counteract this, we can add an additional ```if``` statement that adds to the order book only when the event has been triggered _and_ if the companies shares are not currently in the built-in Quantuopian function ```get_open_orders()```.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def initialize(context):\n",
    "    context.aapl = sid(24)  # Apple share prices\n",
    "    \n",
    "    \n",
    "def handle_data(context, data):  # Runs every minute\n",
    "    hist = data.history(context.aapl, 'price', 50, '1d')\n",
    "    log.info(hist.head())\n",
    "    sma_50 = hist.mean()  # 50 day moving average\n",
    "    sma_20 = hist[-20:].mean()  # 20 day MA\n",
    "    \n",
    "    open_orders = get_open_orders()\n",
    "        \n",
    "    if sma_20 > sma_50:  # If the 20 day moving average has crossed over the 50 day moving average\n",
    "        if context.aapl not in open_orders:\n",
    "            order_target_percent(context.aapl, 1.0)  # Want 100% (1.0) of portfolio to be Apple (long order)\n",
    "    elif sma_50 > sma_20:\n",
    "        if context.aapl not in open_orders:\n",
    "            order_target_percent(context.aapl, -1.0)  # Short order on Apple\n",
    "        \n",
    "    record(leverage=context.account.leverage)  # Record leverage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"sched\"></a> Schedule Function \n",
    "\n",
    "In practice we want to reduce the number of trades we make in a day, such that to avoid transaction fees. The ```schedule_function()```  allows you to schedule functions to occur daily/ weekly/ monthly etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Limits the function ma_cross_over_handling to occur daily, \n",
    "# 1 hour after the market opens.\n",
    "def initialize(context):\n",
    "    context.aapl = sid(24)  # Apple share prices\n",
    "    schedule_function(ma_crossover_handling, date_rules.every_day(), \n",
    "                      time_rules.market_open(hours=1)) \n",
    "\n",
    "# Buy/ sell when moving averages have crossed \n",
    "def ma_crossover_handling(context, data):\n",
    "    hist = data.history(context.aapl, 'price', 50, '1d')\n",
    "    sma_50 = hist.mean()  # 50 day moving average\n",
    "    sma_20 = hist[-20:].mean()  # 20 day MA\n",
    "    \n",
    "    open_orders = get_open_orders()\n",
    "        \n",
    "    if sma_20 > sma_50:  # If the 20 day moving average has crossed over the 50 day moving average\n",
    "        if context.aapl not in open_orders:\n",
    "            order_target_percent(context.aapl, 1.0)  # Want 100% (1.0) of portfolio to be Apple (long order)\n",
    "    elif sma_50 > sma_20:\n",
    "        if context.aapl not in open_orders:\n",
    "            order_target_percent(context.aapl, -1.0)  # Short order on Apple\n",
    "        \n",
    "    record(leverage=context.account.leverage)  # Record leverage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"res\" id=\"sched\"></a> Research\n",
    "\n",
    "In this section we will dive into the [research](https://www.quantopian.com/research) environment of Quantopian. \n",
    "\n",
    "Later in this notebook, we will be uinsg [Alphalens](https://www.quantopian.com/posts/alphalens-a-new-tool-for-analyzing-alpha-factors), a tool that allows us to analyze [alpha factors](https://www.investopedia.com/terms/a/alpha.asp). Alpha factors are used in finance to gauge the performance of an investment against a benchmark (e.g. a market index) which is considered to represent the market's movement as a whole.\n",
    "\n",
    "## <a class=\"anchor\" id=\"pipe\"></a> Pipeline\n",
    "\n",
    "[Pipeline](https://www.quantopian.com/tutorials/pipeline) offers a quick and succinct way to whittle away datasets. The main attraction is that it is ___a lot quicker___ than running functions on datasets. This is because pipeline often does not return the dataframe itself during manipulation, but still allows for you to (much more efficiently) apply functions on them.  You ideally want to do everything you can inside pipeline, as doing otherwise can take 5-10 times longer.\n",
    "\n",
    "Pipeline allows us to build the pipeline data that we will use in our trading strategy, which often forms the crux of our calculations in our trading strategies. \n",
    "\n",
    "Using pipeline, we can return the data (in this case sentiment data) and then use alphalens to analyze the alpha factor, which will be discussed in the next section."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from quantopian.pipeline.data.sentdex import sentiment\n",
    "\n",
    "# Generates the sentiment factor over a provided trading period\n",
    "def make_pipeline():\n",
    "    sentiment_factor = sentiment.sentiment_signal.latest\n",
    "    universe = (Q1500US() & sentiment_factor.notnull())\n",
    "    pipe = Pipeline(columns={'sentiment': sentiment_factor,\n",
    "                            'longs': (sentiment_factor>=4),\n",
    "                            'shorts': (sentiment_factor<=-2)},\n",
    "                   screen=universe)\n",
    "    \n",
    "    return pipe\n",
    "\n",
    "result = run_pipeline(make_pipeline(), start_date='2015-05-01', end_date='2016-01-01')\n",
    "\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"alpha\"></a> Analyzing Alpha Factors with Alphalens\n",
    "\n",
    "At the simplest level, '_[alpha](https://www.investopedia.com/terms/a/alpha.asp)_' is a measure of your returns irrespective of the market gain. As a trader, you want alpha to be __high__, as you want to compare your trading strategies to simply holding very common low risk funds such as the S&P 500 index. Our __optimization objective__ therefore, as traders is: \n",
    "\n",
    "$$\\max[\\alpha].$$\n",
    "\n",
    "Any time that you decide to do trading, you introduce risk. Holding some low risk funds such as the S&P 500 index in addition provides you dividends, whereas in trading you will not get those benefits. Therefore, when making the decision to trade it is very important to do it for the right reasons. Sometimes trading may be done in attempt to hedge against the market, and decrease your exposure to its volatility. In most cases however, the appeal of trading is the ideal of being able to outperform and beat the gains of the market as a whole. A measure of this is _alpha_, and is in summary something that you want to be as high as possible - as it indicates that the strategy you have devised is worthwhile, and that you are indeed beating the market.\n",
    "\n",
    "[Beta](https://www.investopedia.com/terms/b/beta.asp) on the other hand, is a measure of the ___[volatility](https://www.investopedia.com/terms/v/volatility.asp)___, or systematic __risk__ of security/ portfolio compared to the market as a whole. Ideally, you want beta to be __neutral__ (i.e. $\\beta = 0$). The value of beta for a given security can indicate the following: \n",
    "* $\\beta = 1$ $:$ The security's price moves with the market\n",
    "* $\\beta < 1$ $:$ The security is theoretically ___less volatile___ than the market\n",
    "* $\\beta > 1$ $:$ The security is theoretically more volatile than with the market\n",
    "\n",
    "To clarify further, if a security has a $\\beta = 1.3$, then it is theoretically 30% _more_ volatile than the market. If a security has a $\\beta = 0.9$, then it is theoretically 10% _less_ volatile than the market.\n",
    "\n",
    "In cases where beta is negative $\\beta < 0$, then the security can be said to move directly opposite to the market (i.e. if the market goes up, then the security goes down). Stocks that have negative beta can therefore be valuable to trade when searching for hedging strategies.\n",
    "\n",
    "In competitions, Quantopian likes to see beta between: \n",
    "\n",
    "$$-0.3 < \\beta < 0.3.$$\n",
    "\n",
    "Having discussed these two fundamental terms, alpha and beta, we will use [alphalens](http://quantopian.github.io/alphalens/) to measure alpha and make efforts to maximize it for different stocks.\n",
    "\n",
    "With alphalens, your objective is to compare your factor (e.g. sentiment factor) over the entire trading period to the market returns of all the assets you might be interested in, and generate alpha there.  You ideally want to __abstract away__ as much as possible __your trading strategy__. You don't really want to check alpha for your strategy so much, as you want to __measure the alpha__ of your __signal__.  Just because you run a backtest and make money with your trading strategy doesn't really mean that the signal you were trading on actually had any alpha. In other words you could have just got lucky. It therefore takes a few backtests to measure the alpha of the assets you are trading, and confirm this as a worthwhile signal. In addtion, if you ran a backtest and you didn't make money - it also doesn't mean that your signal didn't have alpha, it just meant your strategy was bad. Hence, this is why it is so important to measure alpha - which we can achieve using alphalens.\n",
    "\n",
    "First, let us check prices throughout the entire trading period (2015-2016) of a given signal. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Collect share prices throughout entire trading period of the signal\n",
    "assets = result.index.levels[1].unique()\n",
    "pricing = get_pricing(assets, start_date='2014-12-01', end_date='2016-02-01', fields='open_price')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We set the start date to one month previous, and end date 1 month prior. This is so that we have a little data prior, and after our signal. This is because alphalens is going to look a little bit prior, and after to continue computing market returns - so that we can actually look at what was pricing doing before and after the event of our signal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import alphalens\n",
    "\n",
    "# Quantize the sentiment factor into 2 bins (good/ bad sentiment)\n",
    "factor_data = alphalens.utils.get_clean_factor_and_forward_returns(\n",
    "    factor=result.sentiment, \n",
    "    prices=pricing,\n",
    "    quantiles=2,\n",
    ")\n",
    "\n",
    "# Generate informative statistics + visualizations about the factor\n",
    "alphalens.tears.create_full_tear_sheet(\n",
    "    factor_data\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The option ```quantiles``` takes our sentiment data and seperates it into (in this case) 2 groups (_quantiles_). These groups before separation are meant to come from a factor (signal) that is going to range from low to high.  \n",
    "\n",
    "Here we observe from the visualizations and tables produced by alphalens' tear sheet, that ```sentiment``` is an _okay_ alpha factor. In a strategy, we probably would want a stop loss. Rather than just sentiment, it would be more beneficial to seek additional alpha factors - and hopefully ___combine multiple alpha factors___ before running a backtest.  \n",
    "\n",
    "The research of alpha factors, and combining them is generally one of the core challenges of Quantopian and in devising a good strategy. \n",
    "\n",
    "## <a class=\"anchor\" id=\"backtest_alpha\"></a> Backtesting Alpha Factor\n",
    "\n",
    "Since we already have some alpha with our sentiment alpha factor, we will backtest our pipeline that makes long orders if the sentiment factor is high (>=4) and shorts if the sentiment is low (<=-2).\n",
    "\n",
    "Using the sentiment factor that we investigated, here we are going to be __trading__ against it in an actual algorithm.\n",
    "\n",
    "By __splitting half__ of our investment money towards __long__ orders, and half of it to __short__ orders, we make an effort to (hopefully) always maintain a __near zero beta__."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Redistribute pool of investment money towards long and short orders \n",
    "# to get (near) zero beta\n",
    "def rebalance(context, data):\n",
    "    long_secs = context.output[context.output['longs']].index  # Get ticker\n",
    "    long_weight = 0.5 / len(long_secs)\n",
    "    \n",
    "    short_secs = context.output[context.output['shorts']].index\n",
    "    short_weight = -0.5 / len(short_secs)\n",
    "    \n",
    "    for security in long_secs:  # Long security\n",
    "        if data.can_trade(security):  # If we can trade that security\n",
    "            order_target_percent(security, long_weight)\n",
    "            \n",
    "    for security in short_secs:  # Short security\n",
    "        if data.can_trade(security):  # If we can trade that security\n",
    "            order_target_percent(security, short_weight)\n",
    "    \n",
    "    # Iterate over positions that we have, check if the ticker exists, whether we \n",
    "    # can trade it, and if it does not exist in our long or short securities list.\n",
    "    for security in context.portfolio.positions:\n",
    "        if data.can_trade(security) and security not in long_secs and security not in short_secs:\n",
    "            order_target_percent(security,0)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note: When building algorithms to trade securities and investigate alpha factors, it can be helpful to at first __remove the commision fee__ from your program. This is useful as a first exploratory step, to simply see if there is any signal worth further investigating. After a worthwhile amount of alpha is confirmed, you can then make the algorithm more realistic by including a commission fee, and further develop your algorithm and trading strategy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"pyfolio\"></a> Analyzing Backtest with Pyfolio\n",
    "\n",
    "In this section we will take the backtest that we ran of the above trading algorithm, go back to the research environment, and run the backtest instead through __[pyfolio](https://quantopian.github.io/pyfolio/)__, which will give us a little bit more interesting information regarding the backtest. \n",
    "\n",
    "Immediately after finishing your backtest, observe the hyperlink of the current website you are on. In my case, this is\n",
    "\n",
    "https://www.quantopian.com/algorithms/5b62eed1e100bb3a4f075eef/5b63b1631834024239be37ee.\n",
    "\n",
    "Take note of the last string of characters before the forward-slash in the above hyperlink: '5b63b1631834024239be37ee'. We can collect the results of this backtest using Quantopian's built-in ```get_backtest()``` function, where we provide this string of characters as input:  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://www.quantopian.com/algorithms/5b62eed1e100bb3a4f075eef/5b63b1631834024239be37ee\n",
    "bt = get_backtest('5b63b1631834024239be37ee')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can run some descriptive statistics and visualizations by creating a full tear sheet for this backtest data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate descriptive statistics + visualizations of backtest data\n",
    "bt.create_full_tear_sheet()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"strategizing\"></a> Strategizing\n",
    "\n",
    "In previous sections we have covered the beginning of a very general Quantopian workflow template which involved:\n",
    "1. __Universe Selection__: defining the universe of tradeable components (Q1500US - a set of really easy liquid companies on Quantopian)\n",
    "2. __Single Alpha Factor Modeling__: defining and evaluating individual expressions which rank the cross section of equities in your universe (in our case, this was sentiment)\n",
    "\n",
    "Moving forward we will expand upon this workflow, which involves:\n",
    "\n",
    "3. __Alpha Combination__: combine many single alphas into a final alpha which has stronger prediction power than the best single alpha.\n",
    "4. __Risk Model__ define and calculate the set of risk factors (e.g. [drawdown](https://www.investopedia.com/terms/d/drawdown.asp), [volatility](https://www.investopedia.com/terms/v/volatility.asp), [Sharpe ratio](https://www.investopedia.com/terms/s/sharperatio.asp), [Sortino ratio](https://www.investopedia.com/terms/s/sortinoratio.asp)) you want to use to constrain your portfolio.\n",
    "5. __Portfolio Construction__: implement a process which takes your final combined alpha and your risk model to produce a target portfolio that minimizes risk under your model.\n",
    "6. __Execution__: implement a trading process to transition the current portfolio to the target portfolio.\n",
    "\n",
    "## <a class=\"anchor\" id=\"findalpha\"></a> Finding more Alpha Factors\n",
    "\n",
    "Here we will now work on __Alpha Combination__. \n",
    "\n",
    "Our last [strategy](https://www.quantopian.com/algorithms/5b62eed1e100bb3a4f075eef) involved making trading decision based purely on predefined  discrete quantiles for the sentiment factor (```if sentiment >= 4: buy; elif sentiment <= -2: sell```). \n",
    "\n",
    "Here we will remove comission fees, combine alpha factors and then finally improve the method in which we constructed our portfolio while we run a backtest.\n",
    "\n",
    "As our other alpha factor to combine, we will investigate __revenue growth__."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from quantopian.pipeline import Pipeline\n",
    "from quantopian.research import run_pipeline\n",
    "from quantopian.pipeline.filters.morningstar import Q1500US\n",
    "from quantopian.pipeline.data.sentdex import sentiment\n",
    "from quantopian.pipeline.data.morningstar import operation_ratios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build the pipeline, using revenue growth as the new alpha factor\n",
    "def make_pipeline():\n",
    "    testing_factor = operation_ratios.revenue_growth.latest\n",
    "    universe = (Q1500US() & \n",
    "               testing_factor.notnull())\n",
    "    testing_factor = testing_factor.rank(mask=universe, method='average')\n",
    "    \n",
    "    pipe = Pipeline(columns={'testing_factor':testing_factor},\n",
    "                   screen=universe)\n",
    "    \n",
    "    return pipe\n",
    "\n",
    "result = run_pipeline(make_pipeline(), start_date='2015-01-01', end_date='2016-01-01')\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pricing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the full tear sheet of the pricing data\n",
    "import alphalens\n",
    "\n",
    "factor_data = alphalens.utils.get_clean_factor_and_forward_returns(factor = result['testing_factor'],\n",
    "                                        prices = pricing,\n",
    "                                        quantiles = 2,\n",
    "                                        periods = (3,10,30))\n",
    "\n",
    "alphalens.tears.create_full_tear_sheet(factor_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"combinealpha\"></a> Combining Alphas\n",
    "\n",
    "The alpha factors we combine in our pipeline, in our [Quantopian notebook](https://www.quantopian.com/research/notebooks/Combine%20Alphas.ipynb), will be:\n",
    "* [Revenue Growth](https://www.investopedia.com/terms/q/quarterlyrevenuegrowth.asp)\n",
    "* [Operation margin](https://www.investopedia.com/terms/o/operatingmargin.asp)\n",
    "* [Sentiment](https://www.investopedia.com/terms/m/marketsentiment.asp)\n",
    "\n",
    "We combine them into our pipeline as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build the pipeline\n",
    "def make_pipeline():\n",
    "    # Get alpha factors: revenue growth, operation margin, sentiment \n",
    "    testing_factor1 = operation_ratios.revenue_growth.latest\n",
    "    testing_factor2 = operation_ratios.operation_margin.latest\n",
    "    testing_factor3 = sentiment.sentiment_signal.latest\n",
    "    \n",
    "    universe = (Q1500US() & \n",
    "               testing_factor1.notnull() &\n",
    "               testing_factor2.notnull() &\n",
    "               testing_factor3.notnull())\n",
    "    \n",
    "    # Rank alphas\n",
    "    testing_factor1 = testing_factor1.rank(mask=universe, method='average')\n",
    "    testing_factor2 = testing_factor2.rank(mask=universe, method='average')\n",
    "    testing_factor3 = testing_factor3.rank(mask=universe, method='average')\n",
    "\n",
    "    # Combine alphas\n",
    "    testing_factor = testing_factor1 + testing_factor2 + testing_factor3\n",
    " \n",
    "    testing_quantiles = testing_factor.quantiles(2)\n",
    "\n",
    "    pipe = Pipeline(columns={\n",
    "                'testing_factor':testing_factor,\n",
    "                'shorts':testing_quantiles.eq(0),\n",
    "                'longs':testing_quantiles.eq(1)},\n",
    "                    \n",
    "                   screen=universe)\n",
    "    \n",
    "    return pipe\n",
    "\n",
    "result = run_pipeline(make_pipeline(), start_date='2015-01-01', end_date='2016-01-01')\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get pricing data\n",
    "assets = result.index.levels[1].unique()\n",
    "pricing = get_pricing(assets, start_date='2014-12-01', end_date='2016-02-01', fields='open_price')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the full tear sheet of the pricing data, using alphalens\n",
    "factor_data = alphalens.utils.get_clean_factor_and_forward_returns(factor = result['testing_factor'],\n",
    "                                        prices = pricing,\n",
    "                                        quantiles = 2,\n",
    "                                        periods = (3,10,30))\n",
    "\n",
    "alphalens.tears.create_full_tear_sheet(factor_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By observing the output of the [full-tear-sheet](https://www.quantopian.com/posts/factor-tear-sheet), we see that alpha has been improved significantly, and that our beta is within the acceptable range of $0.3 < \\beta < 0.3$. Additionally, the [Information Coefficient](https://www.investopedia.com/terms/i/information-coefficient.asp) (IC) has been been improved.\n",
    "\n",
    "The next step in our [general workflow](#strategizing) template in Quantopian is the Risk Model, would be to constrain our model such that to (for example) limit leverage to a value of 1, and to aim to have low volatility in our assets.\n",
    "\n",
    "In our next step we will discuss Portfolio Construction.\n",
    "\n",
    "## <a class=\"anchor\" id=\"portopt\"></a>  Portfolio Optimization API\n",
    "\n",
    "Scott Sanderson offers an excellent [blog post](https://www.quantopian.com/posts/optimize-api-now-available-in-algorithms) on Quantopian, which discusses the Optimize API in detail.\n",
    "\n",
    "The Optimize API makes use of what is known as [convex optimization](https://en.wikipedia.org/wiki/Convex_optimization). This lets users to specify the desired state of their portfolio in terms of high-level objectives (which we wish to maximize/ minimize) and constraints that the portfolio should adhere to. In this case, convex optimization can be used to (for example) maximize alpha, maximize Sharpe ratio, and/or minimize volatility. Some constraints we may wish to impose on our model could be setting leverage to 1. \n",
    "\n",
    "Below, we merge our approach in combining several alpha factors with the portfolio optimize API approach discused in the previously mentioned [blog post](https://www.quantopian.com/posts/optimize-api-now-available-in-algorithms), borrowing most of their code - but adapting it to use our alpha factors and data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "import quantopian.algorithm as algo\n",
    "import quantopian.experimental.optimize as opt\n",
    "\n",
    "from quantopian.pipeline import Pipeline\n",
    "from quantopian.pipeline.data import builtin, morningstar as mstar\n",
    "from quantopian.pipeline.factors import AverageDollarVolume\n",
    "from quantopian.pipeline.factors.morningstar import MarketCap\n",
    "from quantopian.pipeline.classifiers.morningstar import Sector\n",
    "\n",
    "from quantopian.pipeline.data.sentdex import sentiment\n",
    "from quantopian.pipeline.data.morningstar import operation_ratios\n",
    "from quantopian.pipeline.filters.morningstar import Q1500US\n",
    "\n",
    "# Algorithm Parameters\n",
    "# --------------------\n",
    "# Universe Selection Parameters\n",
    "UNIVERSE_SIZE = 500\n",
    "MIN_MARKET_CAP_PERCENTILE = 50\n",
    "LIQUIDITY_LOOKBACK_LENGTH = 100\n",
    "\n",
    "# Constraint Parameters\n",
    "MAX_GROSS_LEVERAGE = 1.0\n",
    "MAX_SHORT_POSITION_SIZE = 0.002  # 0.02%\n",
    "MAX_LONG_POSITION_SIZE = 0.002   # 0.02%\n",
    "\n",
    "# Scheduling Parameters\n",
    "MINUTES_AFTER_OPEN_TO_TRADE = 10\n",
    "BASE_UNIVERSE_RECALCULATE_FREQUENCY = 'month_start'  # {week,quarter,year}_start are also valid\n",
    "\n",
    "\n",
    "def initialize(context):\n",
    "    # Get alpha factors: revenue growth, operation margin, sentiment \n",
    "    testing_factor1 = operation_ratios.revenue_growth.latest\n",
    "    testing_factor2 = operation_ratios.operation_margin.latest\n",
    "    testing_factor3 = sentiment.sentiment_signal.latest\n",
    "    \n",
    "    universe = (Q1500US() & \n",
    "               testing_factor1.notnull() &\n",
    "               testing_factor2.notnull() &\n",
    "               testing_factor3.notnull())\n",
    "    \n",
    "    # Rank alphas\n",
    "    testing_factor1 = testing_factor1.rank(mask=universe, method='average')\n",
    "    testing_factor2 = testing_factor2.rank(mask=universe, method='average')\n",
    "    testing_factor3 = testing_factor3.rank(mask=universe, method='average')\n",
    "\n",
    "    # Combine alphas\n",
    "    combined_alpha = testing_factor1 + testing_factor2 + testing_factor3\n",
    " \n",
    "\n",
    "    # Schedule Tasks\n",
    "    # --------------\n",
    "    # Create and register a pipeline computing our combined alpha and a sector\n",
    "    # code for every stock in our universe. We'll use these values in our \n",
    "    # optimization below.\n",
    "    pipe = Pipeline(\n",
    "        columns={\n",
    "            'alpha': combined_alpha,\n",
    "            'sector': Sector(),\n",
    "        },\n",
    "        # combined_alpha will be NaN for all stocks not in our universe,\n",
    "        # but we also want to make sure that we have a sector code for everything\n",
    "        # we trade.\n",
    "        screen=combined_alpha.notnull() & Sector().notnull(),\n",
    "    )\n",
    "    algo.attach_pipeline(pipe, 'pipe')\n",
    "\n",
    "    # Schedule a function, 'do_portfolio_construction', to run once a week\n",
    "    # ten minutes after market open.\n",
    "    algo.schedule_function(\n",
    "        do_portfolio_construction,\n",
    "        date_rule=algo.date_rules.week_start(),\n",
    "        time_rule=algo.time_rules.market_open(minutes=MINUTES_AFTER_OPEN_TO_TRADE),\n",
    "        half_days=False,\n",
    "    )\n",
    "\n",
    "\n",
    "def before_trading_start(context, data):\n",
    "    # Call pipeline_output in before_trading_start so that pipeline\n",
    "    # computations happen in the 5 minute timeout of BTS instead of the 1\n",
    "    # minute timeout of handle_data/scheduled functions.\n",
    "    context.pipeline_data = algo.pipeline_output('pipe')\n",
    "\n",
    "\n",
    "# Portfolio Construction\n",
    "# ----------------------\n",
    "def do_portfolio_construction(context, data):\n",
    "    pipeline_data = context.pipeline_data\n",
    "    todays_universe = pipeline_data.index\n",
    "\n",
    "    # Objective\n",
    "    # ---------\n",
    "    # For our objective, we simply use our naive ranks as an alpha coefficient\n",
    "    # and try to maximize that alpha.\n",
    "    # \n",
    "    # This is a **very** naive model. Since our alphas are so widely spread out,\n",
    "    # we should expect to always allocate the maximum amount of long/short\n",
    "    # capital to assets with high/low ranks.\n",
    "    #\n",
    "    # A more sophisticated model would apply some re-scaling here to try to generate\n",
    "    # more meaningful predictions of future returns.\n",
    "    objective = opt.MaximizeAlpha(pipeline_data.alpha)\n",
    "\n",
    "    # Constraints\n",
    "    # -----------\n",
    "    # Constrain our gross leverage to 1.0 or less. This means that the absolute\n",
    "    # value of our long and short positions should not exceed the value of our\n",
    "    # portfolio.\n",
    "    constrain_gross_leverage = opt.MaxGrossLeverage(MAX_GROSS_LEVERAGE)\n",
    "    \n",
    "    # Constrain individual position size to no more than a fixed percentage \n",
    "    # of our portfolio. Because our alphas are so widely distributed, we \n",
    "    # should expect to end up hitting this max for every stock in our universe.\n",
    "    constrain_pos_size = opt.PositionConcentration.with_equal_bounds(\n",
    "        -MAX_SHORT_POSITION_SIZE,\n",
    "        MAX_LONG_POSITION_SIZE,\n",
    "    )\n",
    "\n",
    "    # Constrain ourselves to allocate the same amount of capital to \n",
    "    # long and short positions.\n",
    "    market_neutral = opt.DollarNeutral()\n",
    "    \n",
    "    # Constrain ourselve to have a net leverage of 0.0 in each sector.\n",
    "    sector_neutral = opt.NetPartitionExposure.with_equal_bounds(\n",
    "        labels=pipeline_data.sector,\n",
    "        min=-0.0001,\n",
    "        max=0.0001,\n",
    "    )\n",
    "\n",
    "    # Run the optimization. This will calculate new portfolio weights and\n",
    "    # manage moving our portfolio toward the target.\n",
    "    algo.order_optimal_portfolio(\n",
    "        objective=objective,\n",
    "        constraints=[\n",
    "            constrain_gross_leverage,\n",
    "            constrain_pos_size,\n",
    "            market_neutral,\n",
    "            sector_neutral,\n",
    "        ],\n",
    "        universe=todays_universe,\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the last string of characters in the hyperlink of our [backtest](https://www.quantopian.com/algorithms/5b680421372f590044cb2eb7/5b680e541d459d43dfd41250), we can collect the backtest in Quantopian's iPython [notebook](https://www.quantopian.com/research/notebooks/Untitled1.ipynb). Having done this, we can create descriptive statistics and visualizations of the backtest using a full tear sheet."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bt = get_backtest('5b680e541d459d43dfd41250')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bt.create_full_tear_sheet()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we observe that we have made huge improvements to our trading strategy just by combining and trading off weak alpha factors (in this case: sentiment, revenue growth, and operation margin)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <a class=\"anchor\" id=\"localback\"></a> Zipline\n",
    "\n",
    "[Zipline](https://github.com/quantopian/zipline) is arguably the best open-sourced library for __back-testing__. In addition, Zipline is the back-end that runs on Quantopian.\n",
    "\n",
    "In this section we will be using Zipline _locally_, as opposed to using Quantopian's web interface.\n",
    "\n",
    "With Zipline you can backtest using many different assets and currencies, including Bitcoin - which we will discuss later.\n",
    "\n",
    "Installing Zipline can be a bit confusing, but Sentdex offers a useful [video tutorial](https://www.youtube.com/watch?v=J4LTIixReMA&index=25&list=PLQVvvaa0QuDcOdF96TBtRtuQksErCEBYZ) walking you through the installation process. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Allows you to use the Zipline module within this notebook\n",
    "%load_ext zipline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from zipline.api import order, record, symbol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def initialize(context):\n",
    "    pass\n",
    "\n",
    "def handle_data(context, data):\n",
    "    order(symbol('AAPL'), 10)\n",
    "    record(AAPL=data.current(symbol('AAPL'), 'price'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>algo_volatility</th>\n",
       "      <th>algorithm_period_return</th>\n",
       "      <th>alpha</th>\n",
       "      <th>benchmark_period_return</th>\n",
       "      <th>benchmark_volatility</th>\n",
       "      <th>beta</th>\n",
       "      <th>capital_used</th>\n",
       "      <th>ending_cash</th>\n",
       "      <th>ending_exposure</th>\n",
       "      <th>...</th>\n",
       "      <th>short_exposure</th>\n",
       "      <th>short_value</th>\n",
       "      <th>shorts_count</th>\n",
       "      <th>sortino</th>\n",
       "      <th>starting_cash</th>\n",
       "      <th>starting_exposure</th>\n",
       "      <th>starting_value</th>\n",
       "      <th>trading_days</th>\n",
       "      <th>transactions</th>\n",
       "      <th>treasury_period_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2008-01-02 21:00:00+00:00</th>\n",
       "      <td>194.840</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.003665</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1</td>\n",
       "      <td>[]</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-03 21:00:00+00:00</th>\n",
       "      <td>194.930</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>-9.846500e-08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.007343</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1950.284650</td>\n",
       "      <td>9.998050e+06</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-11.224972</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2</td>\n",
       "      <td>[{'price': 195.027465, 'order_id': '2802f8c018...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-04 21:00:00+00:00</th>\n",
       "      <td>180.050</td>\n",
       "      <td>0.000137</td>\n",
       "      <td>-1.506949e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.011035</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1801.410250</td>\n",
       "      <td>9.996248e+06</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-9.225231</td>\n",
       "      <td>9.998050e+06</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>3</td>\n",
       "      <td>[{'price': 180.140025, 'order_id': 'f603ddd817...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-07 21:00:00+00:00</th>\n",
       "      <td>177.640</td>\n",
       "      <td>0.000112</td>\n",
       "      <td>-1.997931e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.014741</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1777.298200</td>\n",
       "      <td>9.994471e+06</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-10.064893</td>\n",
       "      <td>9.996248e+06</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>4</td>\n",
       "      <td>[{'price': 177.72882, 'order_id': 'ffee7f5f1af...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-08 21:00:00+00:00</th>\n",
       "      <td>171.250</td>\n",
       "      <td>0.000140</td>\n",
       "      <td>-3.923593e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.018460</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1713.366250</td>\n",
       "      <td>9.992758e+06</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-11.195170</td>\n",
       "      <td>9.994471e+06</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>5</td>\n",
       "      <td>[{'price': 171.335625, 'order_id': '99be2796b4...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-09 21:00:00+00:00</th>\n",
       "      <td>179.400</td>\n",
       "      <td>0.000290</td>\n",
       "      <td>-6.726635e-06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.022192</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1794.907000</td>\n",
       "      <td>9.990963e+06</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.751847</td>\n",
       "      <td>9.992758e+06</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>6</td>\n",
       "      <td>[{'price': 179.4897, 'order_id': '26a631854700...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-10 21:00:00+00:00</th>\n",
       "      <td>178.020</td>\n",
       "      <td>0.000267</td>\n",
       "      <td>-1.371664e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.025939</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1781.100100</td>\n",
       "      <td>9.989182e+06</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.184238</td>\n",
       "      <td>9.990963e+06</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>7</td>\n",
       "      <td>[{'price': 178.10901, 'order_id': 'b5c4b795bbf...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-11 21:00:00+00:00</th>\n",
       "      <td>172.690</td>\n",
       "      <td>0.000299</td>\n",
       "      <td>-4.578399e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.029699</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1727.773450</td>\n",
       "      <td>9.987454e+06</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.239021</td>\n",
       "      <td>9.989182e+06</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>8</td>\n",
       "      <td>[{'price': 172.776345, 'order_id': '83ba09ccc5...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-14 21:00:00+00:00</th>\n",
       "      <td>178.780</td>\n",
       "      <td>0.000379</td>\n",
       "      <td>-3.244380e-06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033473</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1788.703900</td>\n",
       "      <td>9.985665e+06</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.416540</td>\n",
       "      <td>9.987454e+06</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>9</td>\n",
       "      <td>[{'price': 178.86939, 'order_id': '783bf3e154c...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-15 21:00:00+00:00</th>\n",
       "      <td>169.040</td>\n",
       "      <td>0.000529</td>\n",
       "      <td>-8.124990e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.037260</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1691.255200</td>\n",
       "      <td>9.983974e+06</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-4.623698</td>\n",
       "      <td>9.985665e+06</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>10</td>\n",
       "      <td>[{'price': 169.12452, 'order_id': '7603a588085...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-16 21:00:00+00:00</th>\n",
       "      <td>159.640</td>\n",
       "      <td>0.000621</td>\n",
       "      <td>-1.659307e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.041062</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1597.208200</td>\n",
       "      <td>9.982377e+06</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.494960</td>\n",
       "      <td>9.983974e+06</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>11</td>\n",
       "      <td>[{'price': 159.71982000000003, 'order_id': 'e4...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-17 21:00:00+00:00</th>\n",
       "      <td>160.890</td>\n",
       "      <td>0.000606</td>\n",
       "      <td>-1.535122e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.044877</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1609.714450</td>\n",
       "      <td>9.980767e+06</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.752987</td>\n",
       "      <td>9.982377e+06</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>12</td>\n",
       "      <td>[{'price': 160.970445, 'order_id': '11794fdf10...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-18 21:00:00+00:00</th>\n",
       "      <td>161.360</td>\n",
       "      <td>0.000585</td>\n",
       "      <td>-1.484238e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.048707</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1614.416800</td>\n",
       "      <td>9.979153e+06</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.344059</td>\n",
       "      <td>9.980767e+06</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>13</td>\n",
       "      <td>[{'price': 161.44068000000001, 'order_id': '40...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-22 21:00:00+00:00</th>\n",
       "      <td>155.640</td>\n",
       "      <td>0.000613</td>\n",
       "      <td>-2.171427e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.052550</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1557.188200</td>\n",
       "      <td>9.977595e+06</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.567920</td>\n",
       "      <td>9.979153e+06</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>14</td>\n",
       "      <td>[{'price': 155.71782000000002, 'order_id': 'ba...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-23 21:00:00+00:00</th>\n",
       "      <td>139.072</td>\n",
       "      <td>0.001010</td>\n",
       "      <td>-4.325972e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.056408</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1391.425360</td>\n",
       "      <td>9.976204e+06</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.896659</td>\n",
       "      <td>9.977595e+06</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>15</td>\n",
       "      <td>[{'price': 139.141536, 'order_id': 'a9c98d339f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-24 21:00:00+00:00</th>\n",
       "      <td>135.600</td>\n",
       "      <td>0.000979</td>\n",
       "      <td>-4.812740e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.060280</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1356.688000</td>\n",
       "      <td>9.974847e+06</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.299514</td>\n",
       "      <td>9.976204e+06</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>16</td>\n",
       "      <td>[{'price': 135.6678, 'order_id': '70d38e218e49...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-25 21:00:00+00:00</th>\n",
       "      <td>130.009</td>\n",
       "      <td>0.000970</td>\n",
       "      <td>-5.652050e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.064166</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1300.750045</td>\n",
       "      <td>9.973547e+06</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.919264</td>\n",
       "      <td>9.974847e+06</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>17</td>\n",
       "      <td>[{'price': 130.0740045, 'order_id': '01259c426...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-28 21:00:00+00:00</th>\n",
       "      <td>130.009</td>\n",
       "      <td>0.000950</td>\n",
       "      <td>-5.652710e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.068066</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1300.750045</td>\n",
       "      <td>9.972246e+06</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.697040</td>\n",
       "      <td>9.973547e+06</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>18</td>\n",
       "      <td>[{'price': 130.0740045, 'order_id': 'dd5acc73d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-29 21:00:00+00:00</th>\n",
       "      <td>131.540</td>\n",
       "      <td>0.000946</td>\n",
       "      <td>-5.393108e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.071980</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1316.067700</td>\n",
       "      <td>9.970930e+06</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.147567</td>\n",
       "      <td>9.972246e+06</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>19</td>\n",
       "      <td>[{'price': 131.60576999999998, 'order_id': '97...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-30 21:00:00+00:00</th>\n",
       "      <td>132.180</td>\n",
       "      <td>0.000932</td>\n",
       "      <td>-5.278579e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.075909</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1322.470900</td>\n",
       "      <td>9.969607e+06</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.818592</td>\n",
       "      <td>9.970930e+06</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>20</td>\n",
       "      <td>[{'price': 132.24609, 'order_id': '3fd3dc2a462...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-31 21:00:00+00:00</th>\n",
       "      <td>135.360</td>\n",
       "      <td>0.000957</td>\n",
       "      <td>-4.675065e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.079852</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1354.286800</td>\n",
       "      <td>9.968253e+06</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.893207</td>\n",
       "      <td>9.969607e+06</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>21</td>\n",
       "      <td>[{'price': 135.42768, 'order_id': 'b8b32c32b11...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-01 21:00:00+00:00</th>\n",
       "      <td>133.750</td>\n",
       "      <td>0.000934</td>\n",
       "      <td>-4.997744e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.083810</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1338.178750</td>\n",
       "      <td>9.966915e+06</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.113220</td>\n",
       "      <td>9.968253e+06</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>22</td>\n",
       "      <td>[{'price': 133.816875, 'order_id': 'd33536fcc2...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-04 21:00:00+00:00</th>\n",
       "      <td>131.650</td>\n",
       "      <td>0.000915</td>\n",
       "      <td>-5.439412e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.087782</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1317.168250</td>\n",
       "      <td>9.965598e+06</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.425988</td>\n",
       "      <td>9.966915e+06</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>23</td>\n",
       "      <td>[{'price': 131.715825, 'order_id': '7be76df3f4...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-05 21:00:00+00:00</th>\n",
       "      <td>129.360</td>\n",
       "      <td>0.000900</td>\n",
       "      <td>-5.943869e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.091769</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1294.256800</td>\n",
       "      <td>9.964303e+06</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.765445</td>\n",
       "      <td>9.965598e+06</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>24</td>\n",
       "      <td>[{'price': 129.42468000000002, 'order_id': 'ab...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-06 21:00:00+00:00</th>\n",
       "      <td>122.000</td>\n",
       "      <td>0.000993</td>\n",
       "      <td>-7.637289e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.095770</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1220.620000</td>\n",
       "      <td>9.963083e+06</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.320042</td>\n",
       "      <td>9.964303e+06</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>25</td>\n",
       "      <td>[{'price': 122.061, 'order_id': 'a934f3a3e4724...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-07 21:00:00+00:00</th>\n",
       "      <td>121.240</td>\n",
       "      <td>0.000974</td>\n",
       "      <td>-7.820305e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.099786</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1213.016200</td>\n",
       "      <td>9.961870e+06</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.338775</td>\n",
       "      <td>9.963083e+06</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>26</td>\n",
       "      <td>[{'price': 121.30062000000001, 'order_id': '2f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-08 21:00:00+00:00</th>\n",
       "      <td>125.480</td>\n",
       "      <td>0.001042</td>\n",
       "      <td>-6.760943e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.103817</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1255.437400</td>\n",
       "      <td>9.960614e+06</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.225585</td>\n",
       "      <td>9.961870e+06</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>27</td>\n",
       "      <td>[{'price': 125.54274000000001, 'order_id': 'ab...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-11 21:00:00+00:00</th>\n",
       "      <td>129.449</td>\n",
       "      <td>0.001092</td>\n",
       "      <td>-5.729660e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.107862</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1295.147245</td>\n",
       "      <td>9.959319e+06</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.180489</td>\n",
       "      <td>9.960614e+06</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>28</td>\n",
       "      <td>[{'price': 129.51372450000002, 'order_id': '2d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-12 21:00:00+00:00</th>\n",
       "      <td>124.860</td>\n",
       "      <td>0.001115</td>\n",
       "      <td>-6.969324e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.111922</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1249.234300</td>\n",
       "      <td>9.958070e+06</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.800272</td>\n",
       "      <td>9.959319e+06</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>29</td>\n",
       "      <td>[{'price': 124.92243, 'order_id': 'b415841c42f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-13 21:00:00+00:00</th>\n",
       "      <td>129.400</td>\n",
       "      <td>0.001180</td>\n",
       "      <td>-5.698781e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.115998</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1294.657000</td>\n",
       "      <td>9.956775e+06</td>\n",
       "      <td>37526.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-4.662658</td>\n",
       "      <td>9.958070e+06</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>30</td>\n",
       "      <td>[{'price': 129.4647, 'order_id': 'c4c8109ced65...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-17 21:00:00+00:00</th>\n",
       "      <td>377.410</td>\n",
       "      <td>0.042831</td>\n",
       "      <td>1.506754e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.017863</td>\n",
       "      <td>0.005238</td>\n",
       "      <td>-0.575354</td>\n",
       "      <td>-3775.997050</td>\n",
       "      <td>7.811911e+06</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.265484</td>\n",
       "      <td>7.815687e+06</td>\n",
       "      <td>3763050.60</td>\n",
       "      <td>3763050.60</td>\n",
       "      <td>980</td>\n",
       "      <td>[{'price': 377.59870500000005, 'order_id': '6a...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-18 21:00:00+00:00</th>\n",
       "      <td>374.940</td>\n",
       "      <td>0.042825</td>\n",
       "      <td>1.482571e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.146203</td>\n",
       "      <td>0.005235</td>\n",
       "      <td>-0.575979</td>\n",
       "      <td>-3751.284700</td>\n",
       "      <td>7.808159e+06</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.245542</td>\n",
       "      <td>7.811911e+06</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>981</td>\n",
       "      <td>[{'price': 375.12747, 'order_id': 'a78d598a213...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-21 21:00:00+00:00</th>\n",
       "      <td>369.010</td>\n",
       "      <td>0.042884</td>\n",
       "      <td>1.424455e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.275013</td>\n",
       "      <td>0.005232</td>\n",
       "      <td>-0.577433</td>\n",
       "      <td>-3691.955050</td>\n",
       "      <td>7.804467e+06</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.195890</td>\n",
       "      <td>7.808159e+06</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>982</td>\n",
       "      <td>[{'price': 369.194505, 'order_id': '14ef45fb2d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-22 21:00:00+00:00</th>\n",
       "      <td>376.510</td>\n",
       "      <td>0.042981</td>\n",
       "      <td>1.498029e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.404296</td>\n",
       "      <td>0.005230</td>\n",
       "      <td>-0.575665</td>\n",
       "      <td>-3766.992550</td>\n",
       "      <td>7.800700e+06</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.251567</td>\n",
       "      <td>7.804467e+06</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>983</td>\n",
       "      <td>[{'price': 376.698255, 'order_id': '1add0085de...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-23 21:00:00+00:00</th>\n",
       "      <td>366.990</td>\n",
       "      <td>0.043163</td>\n",
       "      <td>1.404540e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.534052</td>\n",
       "      <td>0.005227</td>\n",
       "      <td>-0.577974</td>\n",
       "      <td>-3671.744950</td>\n",
       "      <td>7.797029e+06</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.168451</td>\n",
       "      <td>7.800700e+06</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>984</td>\n",
       "      <td>[{'price': 367.173495, 'order_id': 'd9d979320e...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-25 18:00:00+00:00</th>\n",
       "      <td>363.570</td>\n",
       "      <td>0.043169</td>\n",
       "      <td>1.370920e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.664284</td>\n",
       "      <td>0.005225</td>\n",
       "      <td>-0.578831</td>\n",
       "      <td>-3637.527850</td>\n",
       "      <td>7.793391e+06</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.140921</td>\n",
       "      <td>7.797029e+06</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>985</td>\n",
       "      <td>[{'price': 363.751785, 'order_id': '94e4a4dc0f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-28 21:00:00+00:00</th>\n",
       "      <td>376.120</td>\n",
       "      <td>0.043486</td>\n",
       "      <td>1.494410e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.794994</td>\n",
       "      <td>0.005222</td>\n",
       "      <td>-0.575833</td>\n",
       "      <td>-3763.090600</td>\n",
       "      <td>7.789628e+06</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.234077</td>\n",
       "      <td>7.793391e+06</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>986</td>\n",
       "      <td>[{'price': 376.30806, 'order_id': '26d5e56042d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-29 21:00:00+00:00</th>\n",
       "      <td>373.200</td>\n",
       "      <td>0.043485</td>\n",
       "      <td>1.465646e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.926182</td>\n",
       "      <td>0.005219</td>\n",
       "      <td>-0.576566</td>\n",
       "      <td>-3733.876000</td>\n",
       "      <td>7.785894e+06</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.210759</td>\n",
       "      <td>7.789628e+06</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>987</td>\n",
       "      <td>[{'price': 373.3866, 'order_id': 'dd7d39c74332...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-30 21:00:00+00:00</th>\n",
       "      <td>382.200</td>\n",
       "      <td>0.043632</td>\n",
       "      <td>1.554384e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.057851</td>\n",
       "      <td>0.005217</td>\n",
       "      <td>-0.574446</td>\n",
       "      <td>-3823.921000</td>\n",
       "      <td>7.782070e+06</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.276818</td>\n",
       "      <td>7.785894e+06</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>988</td>\n",
       "      <td>[{'price': 382.3911, 'order_id': 'c8d47ba83a0f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-01 21:00:00+00:00</th>\n",
       "      <td>387.930</td>\n",
       "      <td>0.043675</td>\n",
       "      <td>1.610937e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.190003</td>\n",
       "      <td>0.005214</td>\n",
       "      <td>-0.573122</td>\n",
       "      <td>-3881.249650</td>\n",
       "      <td>7.778189e+06</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.318314</td>\n",
       "      <td>7.782070e+06</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>989</td>\n",
       "      <td>[{'price': 388.123965, 'order_id': '4103ec0d23...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-02 21:00:00+00:00</th>\n",
       "      <td>389.700</td>\n",
       "      <td>0.043659</td>\n",
       "      <td>1.628423e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.322639</td>\n",
       "      <td>0.005211</td>\n",
       "      <td>-0.572742</td>\n",
       "      <td>-3898.958500</td>\n",
       "      <td>7.774290e+06</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.330608</td>\n",
       "      <td>7.778189e+06</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>990</td>\n",
       "      <td>[{'price': 389.89484999999996, 'order_id': 'ef...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-05 21:00:00+00:00</th>\n",
       "      <td>393.010</td>\n",
       "      <td>0.043657</td>\n",
       "      <td>1.661157e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.455761</td>\n",
       "      <td>0.005209</td>\n",
       "      <td>-0.571999</td>\n",
       "      <td>-3932.075050</td>\n",
       "      <td>7.770358e+06</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.354150</td>\n",
       "      <td>7.774290e+06</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>991</td>\n",
       "      <td>[{'price': 393.206505, 'order_id': 'b4ac822284...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-06 21:00:00+00:00</th>\n",
       "      <td>390.950</td>\n",
       "      <td>0.043646</td>\n",
       "      <td>1.640761e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.589371</td>\n",
       "      <td>0.005206</td>\n",
       "      <td>-0.572524</td>\n",
       "      <td>-3911.464750</td>\n",
       "      <td>7.766446e+06</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.337835</td>\n",
       "      <td>7.770358e+06</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>992</td>\n",
       "      <td>[{'price': 391.145475, 'order_id': '487c2534a2...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-07 21:00:00+00:00</th>\n",
       "      <td>389.090</td>\n",
       "      <td>0.043633</td>\n",
       "      <td>1.622326e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.723471</td>\n",
       "      <td>0.005204</td>\n",
       "      <td>-0.573001</td>\n",
       "      <td>-3892.855450</td>\n",
       "      <td>7.762554e+06</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.323077</td>\n",
       "      <td>7.766446e+06</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>993</td>\n",
       "      <td>[{'price': 389.28454500000004, 'order_id': 'ed...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-08 21:00:00+00:00</th>\n",
       "      <td>390.660</td>\n",
       "      <td>0.043615</td>\n",
       "      <td>1.637899e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.858062</td>\n",
       "      <td>0.005201</td>\n",
       "      <td>-0.572668</td>\n",
       "      <td>-3908.563300</td>\n",
       "      <td>7.758645e+06</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.333910</td>\n",
       "      <td>7.762554e+06</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>994</td>\n",
       "      <td>[{'price': 390.85533000000004, 'order_id': 'bb...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-09 21:00:00+00:00</th>\n",
       "      <td>393.620</td>\n",
       "      <td>0.043609</td>\n",
       "      <td>1.667290e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.993146</td>\n",
       "      <td>0.005198</td>\n",
       "      <td>-0.572009</td>\n",
       "      <td>-3938.178100</td>\n",
       "      <td>7.754707e+06</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.354900</td>\n",
       "      <td>7.758645e+06</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>995</td>\n",
       "      <td>[{'price': 393.81681000000003, 'order_id': '6e...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-12 21:00:00+00:00</th>\n",
       "      <td>391.840</td>\n",
       "      <td>0.043595</td>\n",
       "      <td>1.649594e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.128726</td>\n",
       "      <td>0.005196</td>\n",
       "      <td>-0.572467</td>\n",
       "      <td>-3920.369200</td>\n",
       "      <td>7.750786e+06</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.340770</td>\n",
       "      <td>7.754707e+06</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>996</td>\n",
       "      <td>[{'price': 392.03592000000003, 'order_id': 'dd...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-13 21:00:00+00:00</th>\n",
       "      <td>388.810</td>\n",
       "      <td>0.043595</td>\n",
       "      <td>1.619444e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.264802</td>\n",
       "      <td>0.005193</td>\n",
       "      <td>-0.573220</td>\n",
       "      <td>-3890.054050</td>\n",
       "      <td>7.746896e+06</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.316649</td>\n",
       "      <td>7.750786e+06</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>997</td>\n",
       "      <td>[{'price': 389.004405, 'order_id': '8d03848e27...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-14 21:00:00+00:00</th>\n",
       "      <td>380.190</td>\n",
       "      <td>0.043738</td>\n",
       "      <td>1.533587e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.401377</td>\n",
       "      <td>0.005191</td>\n",
       "      <td>-0.575294</td>\n",
       "      <td>-3803.810950</td>\n",
       "      <td>7.743093e+06</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.242998</td>\n",
       "      <td>7.746896e+06</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>998</td>\n",
       "      <td>[{'price': 380.380095, 'order_id': '9ed89feec7...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-15 21:00:00+00:00</th>\n",
       "      <td>378.940</td>\n",
       "      <td>0.043720</td>\n",
       "      <td>1.521122e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.538453</td>\n",
       "      <td>0.005188</td>\n",
       "      <td>-0.575627</td>\n",
       "      <td>-3791.304700</td>\n",
       "      <td>7.739301e+06</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.233004</td>\n",
       "      <td>7.743093e+06</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>999</td>\n",
       "      <td>[{'price': 379.12946999999997, 'order_id': 'b1...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-16 21:00:00+00:00</th>\n",
       "      <td>381.020</td>\n",
       "      <td>0.043706</td>\n",
       "      <td>1.541879e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.676031</td>\n",
       "      <td>0.005186</td>\n",
       "      <td>-0.575167</td>\n",
       "      <td>-3812.115100</td>\n",
       "      <td>7.735489e+06</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.247658</td>\n",
       "      <td>7.739301e+06</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>1000</td>\n",
       "      <td>[{'price': 381.21051, 'order_id': 'd57b785ed86...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-19 21:00:00+00:00</th>\n",
       "      <td>382.210</td>\n",
       "      <td>0.043687</td>\n",
       "      <td>1.553765e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.814113</td>\n",
       "      <td>0.005183</td>\n",
       "      <td>-0.574920</td>\n",
       "      <td>-3824.021050</td>\n",
       "      <td>7.731665e+06</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.255759</td>\n",
       "      <td>7.735489e+06</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>1001</td>\n",
       "      <td>[{'price': 382.401105, 'order_id': '6a1cf5b26a...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-20 21:00:00+00:00</th>\n",
       "      <td>395.950</td>\n",
       "      <td>0.044060</td>\n",
       "      <td>1.691163e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.952702</td>\n",
       "      <td>0.005180</td>\n",
       "      <td>-0.571692</td>\n",
       "      <td>-3961.489750</td>\n",
       "      <td>7.727704e+06</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.355833</td>\n",
       "      <td>7.731665e+06</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>1002</td>\n",
       "      <td>[{'price': 396.147975, 'order_id': 'c42d08b8be...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-21 21:00:00+00:00</th>\n",
       "      <td>396.445</td>\n",
       "      <td>0.044039</td>\n",
       "      <td>1.696116e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.091798</td>\n",
       "      <td>0.005178</td>\n",
       "      <td>-0.571615</td>\n",
       "      <td>-3966.442225</td>\n",
       "      <td>7.723737e+06</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.358742</td>\n",
       "      <td>7.727704e+06</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>1003</td>\n",
       "      <td>[{'price': 396.6432225, 'order_id': '94695baad...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-22 21:00:00+00:00</th>\n",
       "      <td>398.550</td>\n",
       "      <td>0.044024</td>\n",
       "      <td>1.717206e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.231404</td>\n",
       "      <td>0.005175</td>\n",
       "      <td>-0.571160</td>\n",
       "      <td>-3987.502750</td>\n",
       "      <td>7.719750e+06</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.373319</td>\n",
       "      <td>7.723737e+06</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>1004</td>\n",
       "      <td>[{'price': 398.749275, 'order_id': '92debfff54...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-23 21:00:00+00:00</th>\n",
       "      <td>403.330</td>\n",
       "      <td>0.044046</td>\n",
       "      <td>1.765148e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.371522</td>\n",
       "      <td>0.005173</td>\n",
       "      <td>-0.570080</td>\n",
       "      <td>-4035.326650</td>\n",
       "      <td>7.715714e+06</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.407230</td>\n",
       "      <td>7.719750e+06</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>1005</td>\n",
       "      <td>[{'price': 403.531665, 'order_id': '30c49c6213...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-27 21:00:00+00:00</th>\n",
       "      <td>406.530</td>\n",
       "      <td>0.044043</td>\n",
       "      <td>1.797274e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.512153</td>\n",
       "      <td>0.005170</td>\n",
       "      <td>-0.569374</td>\n",
       "      <td>-4067.342650</td>\n",
       "      <td>7.711647e+06</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.429607</td>\n",
       "      <td>7.715714e+06</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>1006</td>\n",
       "      <td>[{'price': 406.733265, 'order_id': '3f153af629...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-28 21:00:00+00:00</th>\n",
       "      <td>402.640</td>\n",
       "      <td>0.044056</td>\n",
       "      <td>1.758177e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.653300</td>\n",
       "      <td>0.005168</td>\n",
       "      <td>-0.570321</td>\n",
       "      <td>-4028.423200</td>\n",
       "      <td>7.707619e+06</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.398716</td>\n",
       "      <td>7.711647e+06</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>1007</td>\n",
       "      <td>[{'price': 402.84132, 'order_id': '58ed60a8ebc...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-29 21:00:00+00:00</th>\n",
       "      <td>405.120</td>\n",
       "      <td>0.044045</td>\n",
       "      <td>1.783124e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.794963</td>\n",
       "      <td>0.005165</td>\n",
       "      <td>-0.569782</td>\n",
       "      <td>-4053.235600</td>\n",
       "      <td>7.703565e+06</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.415906</td>\n",
       "      <td>7.707619e+06</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>1008</td>\n",
       "      <td>[{'price': 405.32256, 'order_id': '21fa35f5323...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-30 21:00:00+00:00</th>\n",
       "      <td>405.000</td>\n",
       "      <td>0.044023</td>\n",
       "      <td>1.781913e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.937147</td>\n",
       "      <td>0.005162</td>\n",
       "      <td>-0.569851</td>\n",
       "      <td>-4052.035000</td>\n",
       "      <td>7.699513e+06</td>\n",
       "      <td>4082400.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.414337</td>\n",
       "      <td>7.703565e+06</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>1009</td>\n",
       "      <td>[{'price': 405.2025, 'order_id': 'c47392309589...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1009 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              AAPL  algo_volatility  algorithm_period_return  \\\n",
       "2008-01-02 21:00:00+00:00  194.840              NaN             0.000000e+00   \n",
       "2008-01-03 21:00:00+00:00  194.930         0.000001            -9.846500e-08   \n",
       "2008-01-04 21:00:00+00:00  180.050         0.000137            -1.506949e-05   \n",
       "2008-01-07 21:00:00+00:00  177.640         0.000112            -1.997931e-05   \n",
       "2008-01-08 21:00:00+00:00  171.250         0.000140            -3.923593e-05   \n",
       "2008-01-09 21:00:00+00:00  179.400         0.000290            -6.726635e-06   \n",
       "2008-01-10 21:00:00+00:00  178.020         0.000267            -1.371664e-05   \n",
       "2008-01-11 21:00:00+00:00  172.690         0.000299            -4.578399e-05   \n",
       "2008-01-14 21:00:00+00:00  178.780         0.000379            -3.244380e-06   \n",
       "2008-01-15 21:00:00+00:00  169.040         0.000529            -8.124990e-05   \n",
       "2008-01-16 21:00:00+00:00  159.640         0.000621            -1.659307e-04   \n",
       "2008-01-17 21:00:00+00:00  160.890         0.000606            -1.535122e-04   \n",
       "2008-01-18 21:00:00+00:00  161.360         0.000585            -1.484238e-04   \n",
       "2008-01-22 21:00:00+00:00  155.640         0.000613            -2.171427e-04   \n",
       "2008-01-23 21:00:00+00:00  139.072         0.001010            -4.325972e-04   \n",
       "2008-01-24 21:00:00+00:00  135.600         0.000979            -4.812740e-04   \n",
       "2008-01-25 21:00:00+00:00  130.009         0.000970            -5.652050e-04   \n",
       "2008-01-28 21:00:00+00:00  130.009         0.000950            -5.652710e-04   \n",
       "2008-01-29 21:00:00+00:00  131.540         0.000946            -5.393108e-04   \n",
       "2008-01-30 21:00:00+00:00  132.180         0.000932            -5.278579e-04   \n",
       "2008-01-31 21:00:00+00:00  135.360         0.000957            -4.675065e-04   \n",
       "2008-02-01 21:00:00+00:00  133.750         0.000934            -4.997744e-04   \n",
       "2008-02-04 21:00:00+00:00  131.650         0.000915            -5.439412e-04   \n",
       "2008-02-05 21:00:00+00:00  129.360         0.000900            -5.943869e-04   \n",
       "2008-02-06 21:00:00+00:00  122.000         0.000993            -7.637289e-04   \n",
       "2008-02-07 21:00:00+00:00  121.240         0.000974            -7.820305e-04   \n",
       "2008-02-08 21:00:00+00:00  125.480         0.001042            -6.760943e-04   \n",
       "2008-02-11 21:00:00+00:00  129.449         0.001092            -5.729660e-04   \n",
       "2008-02-12 21:00:00+00:00  124.860         0.001115            -6.969324e-04   \n",
       "2008-02-13 21:00:00+00:00  129.400         0.001180            -5.698781e-04   \n",
       "...                            ...              ...                      ...   \n",
       "2011-11-17 21:00:00+00:00  377.410         0.042831             1.506754e-01   \n",
       "2011-11-18 21:00:00+00:00  374.940         0.042825             1.482571e-01   \n",
       "2011-11-21 21:00:00+00:00  369.010         0.042884             1.424455e-01   \n",
       "2011-11-22 21:00:00+00:00  376.510         0.042981             1.498029e-01   \n",
       "2011-11-23 21:00:00+00:00  366.990         0.043163             1.404540e-01   \n",
       "2011-11-25 18:00:00+00:00  363.570         0.043169             1.370920e-01   \n",
       "2011-11-28 21:00:00+00:00  376.120         0.043486             1.494410e-01   \n",
       "2011-11-29 21:00:00+00:00  373.200         0.043485             1.465646e-01   \n",
       "2011-11-30 21:00:00+00:00  382.200         0.043632             1.554384e-01   \n",
       "2011-12-01 21:00:00+00:00  387.930         0.043675             1.610937e-01   \n",
       "2011-12-02 21:00:00+00:00  389.700         0.043659             1.628423e-01   \n",
       "2011-12-05 21:00:00+00:00  393.010         0.043657             1.661157e-01   \n",
       "2011-12-06 21:00:00+00:00  390.950         0.043646             1.640761e-01   \n",
       "2011-12-07 21:00:00+00:00  389.090         0.043633             1.622326e-01   \n",
       "2011-12-08 21:00:00+00:00  390.660         0.043615             1.637899e-01   \n",
       "2011-12-09 21:00:00+00:00  393.620         0.043609             1.667290e-01   \n",
       "2011-12-12 21:00:00+00:00  391.840         0.043595             1.649594e-01   \n",
       "2011-12-13 21:00:00+00:00  388.810         0.043595             1.619444e-01   \n",
       "2011-12-14 21:00:00+00:00  380.190         0.043738             1.533587e-01   \n",
       "2011-12-15 21:00:00+00:00  378.940         0.043720             1.521122e-01   \n",
       "2011-12-16 21:00:00+00:00  381.020         0.043706             1.541879e-01   \n",
       "2011-12-19 21:00:00+00:00  382.210         0.043687             1.553765e-01   \n",
       "2011-12-20 21:00:00+00:00  395.950         0.044060             1.691163e-01   \n",
       "2011-12-21 21:00:00+00:00  396.445         0.044039             1.696116e-01   \n",
       "2011-12-22 21:00:00+00:00  398.550         0.044024             1.717206e-01   \n",
       "2011-12-23 21:00:00+00:00  403.330         0.044046             1.765148e-01   \n",
       "2011-12-27 21:00:00+00:00  406.530         0.044043             1.797274e-01   \n",
       "2011-12-28 21:00:00+00:00  402.640         0.044056             1.758177e-01   \n",
       "2011-12-29 21:00:00+00:00  405.120         0.044045             1.783124e-01   \n",
       "2011-12-30 21:00:00+00:00  405.000         0.044023             1.781913e-01   \n",
       "\n",
       "                              alpha  benchmark_period_return  \\\n",
       "2008-01-02 21:00:00+00:00       NaN                 0.003665   \n",
       "2008-01-03 21:00:00+00:00       NaN                 0.007343   \n",
       "2008-01-04 21:00:00+00:00       NaN                 0.011035   \n",
       "2008-01-07 21:00:00+00:00       NaN                 0.014741   \n",
       "2008-01-08 21:00:00+00:00       NaN                 0.018460   \n",
       "2008-01-09 21:00:00+00:00       NaN                 0.022192   \n",
       "2008-01-10 21:00:00+00:00       NaN                 0.025939   \n",
       "2008-01-11 21:00:00+00:00       NaN                 0.029699   \n",
       "2008-01-14 21:00:00+00:00       NaN                 0.033473   \n",
       "2008-01-15 21:00:00+00:00       NaN                 0.037260   \n",
       "2008-01-16 21:00:00+00:00       NaN                 0.041062   \n",
       "2008-01-17 21:00:00+00:00       NaN                 0.044877   \n",
       "2008-01-18 21:00:00+00:00       NaN                 0.048707   \n",
       "2008-01-22 21:00:00+00:00       NaN                 0.052550   \n",
       "2008-01-23 21:00:00+00:00       NaN                 0.056408   \n",
       "2008-01-24 21:00:00+00:00       NaN                 0.060280   \n",
       "2008-01-25 21:00:00+00:00       NaN                 0.064166   \n",
       "2008-01-28 21:00:00+00:00       NaN                 0.068066   \n",
       "2008-01-29 21:00:00+00:00       NaN                 0.071980   \n",
       "2008-01-30 21:00:00+00:00       NaN                 0.075909   \n",
       "2008-01-31 21:00:00+00:00       NaN                 0.079852   \n",
       "2008-02-01 21:00:00+00:00       NaN                 0.083810   \n",
       "2008-02-04 21:00:00+00:00       NaN                 0.087782   \n",
       "2008-02-05 21:00:00+00:00       NaN                 0.091769   \n",
       "2008-02-06 21:00:00+00:00       NaN                 0.095770   \n",
       "2008-02-07 21:00:00+00:00       NaN                 0.099786   \n",
       "2008-02-08 21:00:00+00:00       NaN                 0.103817   \n",
       "2008-02-11 21:00:00+00:00       NaN                 0.107862   \n",
       "2008-02-12 21:00:00+00:00       NaN                 0.111922   \n",
       "2008-02-13 21:00:00+00:00       NaN                 0.115998   \n",
       "...                             ...                      ...   \n",
       "2011-11-17 21:00:00+00:00  0.564055                34.017863   \n",
       "2011-11-18 21:00:00+00:00  0.564055                34.146203   \n",
       "2011-11-21 21:00:00+00:00  0.564055                34.275013   \n",
       "2011-11-22 21:00:00+00:00  0.564055                34.404296   \n",
       "2011-11-23 21:00:00+00:00  0.564055                34.534052   \n",
       "2011-11-25 18:00:00+00:00  0.564055                34.664284   \n",
       "2011-11-28 21:00:00+00:00  0.564055                34.794994   \n",
       "2011-11-29 21:00:00+00:00  0.564055                34.926182   \n",
       "2011-11-30 21:00:00+00:00  0.564055                35.057851   \n",
       "2011-12-01 21:00:00+00:00  0.564055                35.190003   \n",
       "2011-12-02 21:00:00+00:00  0.564055                35.322639   \n",
       "2011-12-05 21:00:00+00:00  0.564055                35.455761   \n",
       "2011-12-06 21:00:00+00:00  0.564055                35.589371   \n",
       "2011-12-07 21:00:00+00:00  0.564055                35.723471   \n",
       "2011-12-08 21:00:00+00:00  0.564055                35.858062   \n",
       "2011-12-09 21:00:00+00:00  0.564055                35.993146   \n",
       "2011-12-12 21:00:00+00:00  0.564055                36.128726   \n",
       "2011-12-13 21:00:00+00:00  0.564055                36.264802   \n",
       "2011-12-14 21:00:00+00:00  0.564055                36.401377   \n",
       "2011-12-15 21:00:00+00:00  0.564055                36.538453   \n",
       "2011-12-16 21:00:00+00:00  0.564055                36.676031   \n",
       "2011-12-19 21:00:00+00:00  0.564055                36.814113   \n",
       "2011-12-20 21:00:00+00:00  0.564055                36.952702   \n",
       "2011-12-21 21:00:00+00:00  0.564055                37.091798   \n",
       "2011-12-22 21:00:00+00:00  0.564055                37.231404   \n",
       "2011-12-23 21:00:00+00:00  0.564055                37.371522   \n",
       "2011-12-27 21:00:00+00:00  0.564055                37.512153   \n",
       "2011-12-28 21:00:00+00:00  0.564055                37.653300   \n",
       "2011-12-29 21:00:00+00:00  0.564055                37.794963   \n",
       "2011-12-30 21:00:00+00:00  0.564055                37.937147   \n",
       "\n",
       "                           benchmark_volatility      beta  capital_used  \\\n",
       "2008-01-02 21:00:00+00:00                   NaN       NaN      0.000000   \n",
       "2008-01-03 21:00:00+00:00              0.000000       NaN  -1950.284650   \n",
       "2008-01-04 21:00:00+00:00              0.000000       NaN  -1801.410250   \n",
       "2008-01-07 21:00:00+00:00              0.000000       NaN  -1777.298200   \n",
       "2008-01-08 21:00:00+00:00              0.000000       NaN  -1713.366250   \n",
       "2008-01-09 21:00:00+00:00              0.000000       NaN  -1794.907000   \n",
       "2008-01-10 21:00:00+00:00              0.000000       NaN  -1781.100100   \n",
       "2008-01-11 21:00:00+00:00              0.000000       NaN  -1727.773450   \n",
       "2008-01-14 21:00:00+00:00              0.000000       NaN  -1788.703900   \n",
       "2008-01-15 21:00:00+00:00              0.000000       NaN  -1691.255200   \n",
       "2008-01-16 21:00:00+00:00              0.000000       NaN  -1597.208200   \n",
       "2008-01-17 21:00:00+00:00              0.000000       NaN  -1609.714450   \n",
       "2008-01-18 21:00:00+00:00              0.000000       NaN  -1614.416800   \n",
       "2008-01-22 21:00:00+00:00              0.000000       NaN  -1557.188200   \n",
       "2008-01-23 21:00:00+00:00              0.000000       NaN  -1391.425360   \n",
       "2008-01-24 21:00:00+00:00              0.000000       NaN  -1356.688000   \n",
       "2008-01-25 21:00:00+00:00              0.000000       NaN  -1300.750045   \n",
       "2008-01-28 21:00:00+00:00              0.000000       NaN  -1300.750045   \n",
       "2008-01-29 21:00:00+00:00              0.000000       NaN  -1316.067700   \n",
       "2008-01-30 21:00:00+00:00              0.000000       NaN  -1322.470900   \n",
       "2008-01-31 21:00:00+00:00              0.000000       NaN  -1354.286800   \n",
       "2008-02-01 21:00:00+00:00              0.000000       NaN  -1338.178750   \n",
       "2008-02-04 21:00:00+00:00              0.000000       NaN  -1317.168250   \n",
       "2008-02-05 21:00:00+00:00              0.000000       NaN  -1294.256800   \n",
       "2008-02-06 21:00:00+00:00              0.000000       NaN  -1220.620000   \n",
       "2008-02-07 21:00:00+00:00              0.000000       NaN  -1213.016200   \n",
       "2008-02-08 21:00:00+00:00              0.000000       NaN  -1255.437400   \n",
       "2008-02-11 21:00:00+00:00              0.000000       NaN  -1295.147245   \n",
       "2008-02-12 21:00:00+00:00              0.000000       NaN  -1249.234300   \n",
       "2008-02-13 21:00:00+00:00              0.000000       NaN  -1294.657000   \n",
       "...                                         ...       ...           ...   \n",
       "2011-11-17 21:00:00+00:00              0.005238 -0.575354  -3775.997050   \n",
       "2011-11-18 21:00:00+00:00              0.005235 -0.575979  -3751.284700   \n",
       "2011-11-21 21:00:00+00:00              0.005232 -0.577433  -3691.955050   \n",
       "2011-11-22 21:00:00+00:00              0.005230 -0.575665  -3766.992550   \n",
       "2011-11-23 21:00:00+00:00              0.005227 -0.577974  -3671.744950   \n",
       "2011-11-25 18:00:00+00:00              0.005225 -0.578831  -3637.527850   \n",
       "2011-11-28 21:00:00+00:00              0.005222 -0.575833  -3763.090600   \n",
       "2011-11-29 21:00:00+00:00              0.005219 -0.576566  -3733.876000   \n",
       "2011-11-30 21:00:00+00:00              0.005217 -0.574446  -3823.921000   \n",
       "2011-12-01 21:00:00+00:00              0.005214 -0.573122  -3881.249650   \n",
       "2011-12-02 21:00:00+00:00              0.005211 -0.572742  -3898.958500   \n",
       "2011-12-05 21:00:00+00:00              0.005209 -0.571999  -3932.075050   \n",
       "2011-12-06 21:00:00+00:00              0.005206 -0.572524  -3911.464750   \n",
       "2011-12-07 21:00:00+00:00              0.005204 -0.573001  -3892.855450   \n",
       "2011-12-08 21:00:00+00:00              0.005201 -0.572668  -3908.563300   \n",
       "2011-12-09 21:00:00+00:00              0.005198 -0.572009  -3938.178100   \n",
       "2011-12-12 21:00:00+00:00              0.005196 -0.572467  -3920.369200   \n",
       "2011-12-13 21:00:00+00:00              0.005193 -0.573220  -3890.054050   \n",
       "2011-12-14 21:00:00+00:00              0.005191 -0.575294  -3803.810950   \n",
       "2011-12-15 21:00:00+00:00              0.005188 -0.575627  -3791.304700   \n",
       "2011-12-16 21:00:00+00:00              0.005186 -0.575167  -3812.115100   \n",
       "2011-12-19 21:00:00+00:00              0.005183 -0.574920  -3824.021050   \n",
       "2011-12-20 21:00:00+00:00              0.005180 -0.571692  -3961.489750   \n",
       "2011-12-21 21:00:00+00:00              0.005178 -0.571615  -3966.442225   \n",
       "2011-12-22 21:00:00+00:00              0.005175 -0.571160  -3987.502750   \n",
       "2011-12-23 21:00:00+00:00              0.005173 -0.570080  -4035.326650   \n",
       "2011-12-27 21:00:00+00:00              0.005170 -0.569374  -4067.342650   \n",
       "2011-12-28 21:00:00+00:00              0.005168 -0.570321  -4028.423200   \n",
       "2011-12-29 21:00:00+00:00              0.005165 -0.569782  -4053.235600   \n",
       "2011-12-30 21:00:00+00:00              0.005162 -0.569851  -4052.035000   \n",
       "\n",
       "                            ending_cash  ending_exposure  \\\n",
       "2008-01-02 21:00:00+00:00  1.000000e+07             0.00   \n",
       "2008-01-03 21:00:00+00:00  9.998050e+06          1949.30   \n",
       "2008-01-04 21:00:00+00:00  9.996248e+06          3601.00   \n",
       "2008-01-07 21:00:00+00:00  9.994471e+06          5329.20   \n",
       "2008-01-08 21:00:00+00:00  9.992758e+06          6850.00   \n",
       "2008-01-09 21:00:00+00:00  9.990963e+06          8970.00   \n",
       "2008-01-10 21:00:00+00:00  9.989182e+06         10681.20   \n",
       "2008-01-11 21:00:00+00:00  9.987454e+06         12088.30   \n",
       "2008-01-14 21:00:00+00:00  9.985665e+06         14302.40   \n",
       "2008-01-15 21:00:00+00:00  9.983974e+06         15213.60   \n",
       "2008-01-16 21:00:00+00:00  9.982377e+06         15964.00   \n",
       "2008-01-17 21:00:00+00:00  9.980767e+06         17697.90   \n",
       "2008-01-18 21:00:00+00:00  9.979153e+06         19363.20   \n",
       "2008-01-22 21:00:00+00:00  9.977595e+06         20233.20   \n",
       "2008-01-23 21:00:00+00:00  9.976204e+06         19470.08   \n",
       "2008-01-24 21:00:00+00:00  9.974847e+06         20340.00   \n",
       "2008-01-25 21:00:00+00:00  9.973547e+06         20801.44   \n",
       "2008-01-28 21:00:00+00:00  9.972246e+06         22101.53   \n",
       "2008-01-29 21:00:00+00:00  9.970930e+06         23677.20   \n",
       "2008-01-30 21:00:00+00:00  9.969607e+06         25114.20   \n",
       "2008-01-31 21:00:00+00:00  9.968253e+06         27072.00   \n",
       "2008-02-01 21:00:00+00:00  9.966915e+06         28087.50   \n",
       "2008-02-04 21:00:00+00:00  9.965598e+06         28963.00   \n",
       "2008-02-05 21:00:00+00:00  9.964303e+06         29752.80   \n",
       "2008-02-06 21:00:00+00:00  9.963083e+06         29280.00   \n",
       "2008-02-07 21:00:00+00:00  9.961870e+06         30310.00   \n",
       "2008-02-08 21:00:00+00:00  9.960614e+06         32624.80   \n",
       "2008-02-11 21:00:00+00:00  9.959319e+06         34951.23   \n",
       "2008-02-12 21:00:00+00:00  9.958070e+06         34960.80   \n",
       "2008-02-13 21:00:00+00:00  9.956775e+06         37526.00   \n",
       "...                                 ...              ...   \n",
       "2011-11-17 21:00:00+00:00  7.811911e+06       3694843.90   \n",
       "2011-11-18 21:00:00+00:00  7.808159e+06       3674412.00   \n",
       "2011-11-21 21:00:00+00:00  7.804467e+06       3619988.10   \n",
       "2011-11-22 21:00:00+00:00  7.800700e+06       3697328.20   \n",
       "2011-11-23 21:00:00+00:00  7.797029e+06       3607511.70   \n",
       "2011-11-25 18:00:00+00:00  7.793391e+06       3577528.80   \n",
       "2011-11-28 21:00:00+00:00  7.789628e+06       3704782.00   \n",
       "2011-11-29 21:00:00+00:00  7.785894e+06       3679752.00   \n",
       "2011-11-30 21:00:00+00:00  7.782070e+06       3772314.00   \n",
       "2011-12-01 21:00:00+00:00  7.778189e+06       3832748.40   \n",
       "2011-12-02 21:00:00+00:00  7.774290e+06       3854133.00   \n",
       "2011-12-05 21:00:00+00:00  7.770358e+06       3890799.00   \n",
       "2011-12-06 21:00:00+00:00  7.766446e+06       3874314.50   \n",
       "2011-12-07 21:00:00+00:00  7.762554e+06       3859772.80   \n",
       "2011-12-08 21:00:00+00:00  7.758645e+06       3879253.80   \n",
       "2011-12-09 21:00:00+00:00  7.754707e+06       3912582.80   \n",
       "2011-12-12 21:00:00+00:00  7.750786e+06       3898808.00   \n",
       "2011-12-13 21:00:00+00:00  7.746896e+06       3872547.60   \n",
       "2011-12-14 21:00:00+00:00  7.743093e+06       3790494.30   \n",
       "2011-12-15 21:00:00+00:00  7.739301e+06       3781821.20   \n",
       "2011-12-16 21:00:00+00:00  7.735489e+06       3806389.80   \n",
       "2011-12-19 21:00:00+00:00  7.731665e+06       3822100.00   \n",
       "2011-12-20 21:00:00+00:00  7.727704e+06       3963459.50   \n",
       "2011-12-21 21:00:00+00:00  7.723737e+06       3972378.90   \n",
       "2011-12-22 21:00:00+00:00  7.719750e+06       3997456.50   \n",
       "2011-12-23 21:00:00+00:00  7.715714e+06       4049433.20   \n",
       "2011-12-27 21:00:00+00:00  7.711647e+06       4085626.50   \n",
       "2011-12-28 21:00:00+00:00  7.707619e+06       4050558.40   \n",
       "2011-12-29 21:00:00+00:00  7.703565e+06       4079558.40   \n",
       "2011-12-30 21:00:00+00:00  7.699513e+06       4082400.00   \n",
       "\n",
       "                                   ...            short_exposure  short_value  \\\n",
       "2008-01-02 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-03 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-04 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-09 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-10 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-11 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-14 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-15 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-16 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-17 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-18 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-24 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-25 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-31 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-01 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-04 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-05 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-06 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-11 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-12 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-13 21:00:00+00:00          ...                       0.0          0.0   \n",
       "...                                ...                       ...          ...   \n",
       "2011-11-17 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-18 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-21 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-25 18:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-01 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-02 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-05 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-06 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-09 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-12 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-13 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-14 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-15 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-16 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-19 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-20 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-21 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-27 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "\n",
       "                           shorts_count    sortino  starting_cash  \\\n",
       "2008-01-02 21:00:00+00:00             0        NaN   1.000000e+07   \n",
       "2008-01-03 21:00:00+00:00             0 -11.224972   1.000000e+07   \n",
       "2008-01-04 21:00:00+00:00             0  -9.225231   9.998050e+06   \n",
       "2008-01-07 21:00:00+00:00             0 -10.064893   9.996248e+06   \n",
       "2008-01-08 21:00:00+00:00             0 -11.195170   9.994471e+06   \n",
       "2008-01-09 21:00:00+00:00             0  -1.751847   9.992758e+06   \n",
       "2008-01-10 21:00:00+00:00             0  -3.184238   9.990963e+06   \n",
       "2008-01-11 21:00:00+00:00             0  -6.239021   9.989182e+06   \n",
       "2008-01-14 21:00:00+00:00             0  -0.416540   9.987454e+06   \n",
       "2008-01-15 21:00:00+00:00             0  -4.623698   9.985665e+06   \n",
       "2008-01-16 21:00:00+00:00             0  -6.494960   9.983974e+06   \n",
       "2008-01-17 21:00:00+00:00             0  -5.752987   9.982377e+06   \n",
       "2008-01-18 21:00:00+00:00             0  -5.344059   9.980767e+06   \n",
       "2008-01-22 21:00:00+00:00             0  -6.567920   9.979153e+06   \n",
       "2008-01-23 21:00:00+00:00             0  -6.896659   9.977595e+06   \n",
       "2008-01-24 21:00:00+00:00             0  -7.299514   9.976204e+06   \n",
       "2008-01-25 21:00:00+00:00             0  -7.919264   9.974847e+06   \n",
       "2008-01-28 21:00:00+00:00             0  -7.697040   9.973547e+06   \n",
       "2008-01-29 21:00:00+00:00             0  -7.147567   9.972246e+06   \n",
       "2008-01-30 21:00:00+00:00             0  -6.818592   9.970930e+06   \n",
       "2008-01-31 21:00:00+00:00             0  -5.893207   9.969607e+06   \n",
       "2008-02-01 21:00:00+00:00             0  -6.113220   9.968253e+06   \n",
       "2008-02-04 21:00:00+00:00             0  -6.425988   9.966915e+06   \n",
       "2008-02-05 21:00:00+00:00             0  -6.765445   9.965598e+06   \n",
       "2008-02-06 21:00:00+00:00             0  -7.320042   9.964303e+06   \n",
       "2008-02-07 21:00:00+00:00             0  -7.338775   9.963083e+06   \n",
       "2008-02-08 21:00:00+00:00             0  -6.225585   9.961870e+06   \n",
       "2008-02-11 21:00:00+00:00             0  -5.180489   9.960614e+06   \n",
       "2008-02-12 21:00:00+00:00             0  -5.800272   9.959319e+06   \n",
       "2008-02-13 21:00:00+00:00             0  -4.662658   9.958070e+06   \n",
       "...                                 ...        ...            ...   \n",
       "2011-11-17 21:00:00+00:00             0   1.265484   7.815687e+06   \n",
       "2011-11-18 21:00:00+00:00             0   1.245542   7.811911e+06   \n",
       "2011-11-21 21:00:00+00:00             0   1.195890   7.808159e+06   \n",
       "2011-11-22 21:00:00+00:00             0   1.251567   7.804467e+06   \n",
       "2011-11-23 21:00:00+00:00             0   1.168451   7.800700e+06   \n",
       "2011-11-25 18:00:00+00:00             0   1.140921   7.797029e+06   \n",
       "2011-11-28 21:00:00+00:00             0   1.234077   7.793391e+06   \n",
       "2011-11-29 21:00:00+00:00             0   1.210759   7.789628e+06   \n",
       "2011-11-30 21:00:00+00:00             0   1.276818   7.785894e+06   \n",
       "2011-12-01 21:00:00+00:00             0   1.318314   7.782070e+06   \n",
       "2011-12-02 21:00:00+00:00             0   1.330608   7.778189e+06   \n",
       "2011-12-05 21:00:00+00:00             0   1.354150   7.774290e+06   \n",
       "2011-12-06 21:00:00+00:00             0   1.337835   7.770358e+06   \n",
       "2011-12-07 21:00:00+00:00             0   1.323077   7.766446e+06   \n",
       "2011-12-08 21:00:00+00:00             0   1.333910   7.762554e+06   \n",
       "2011-12-09 21:00:00+00:00             0   1.354900   7.758645e+06   \n",
       "2011-12-12 21:00:00+00:00             0   1.340770   7.754707e+06   \n",
       "2011-12-13 21:00:00+00:00             0   1.316649   7.750786e+06   \n",
       "2011-12-14 21:00:00+00:00             0   1.242998   7.746896e+06   \n",
       "2011-12-15 21:00:00+00:00             0   1.233004   7.743093e+06   \n",
       "2011-12-16 21:00:00+00:00             0   1.247658   7.739301e+06   \n",
       "2011-12-19 21:00:00+00:00             0   1.255759   7.735489e+06   \n",
       "2011-12-20 21:00:00+00:00             0   1.355833   7.731665e+06   \n",
       "2011-12-21 21:00:00+00:00             0   1.358742   7.727704e+06   \n",
       "2011-12-22 21:00:00+00:00             0   1.373319   7.723737e+06   \n",
       "2011-12-23 21:00:00+00:00             0   1.407230   7.719750e+06   \n",
       "2011-12-27 21:00:00+00:00             0   1.429607   7.715714e+06   \n",
       "2011-12-28 21:00:00+00:00             0   1.398716   7.711647e+06   \n",
       "2011-12-29 21:00:00+00:00             0   1.415906   7.707619e+06   \n",
       "2011-12-30 21:00:00+00:00             0   1.414337   7.703565e+06   \n",
       "\n",
       "                           starting_exposure  starting_value  trading_days  \\\n",
       "2008-01-02 21:00:00+00:00               0.00            0.00             1   \n",
       "2008-01-03 21:00:00+00:00               0.00            0.00             2   \n",
       "2008-01-04 21:00:00+00:00            1949.30         1949.30             3   \n",
       "2008-01-07 21:00:00+00:00            3601.00         3601.00             4   \n",
       "2008-01-08 21:00:00+00:00            5329.20         5329.20             5   \n",
       "2008-01-09 21:00:00+00:00            6850.00         6850.00             6   \n",
       "2008-01-10 21:00:00+00:00            8970.00         8970.00             7   \n",
       "2008-01-11 21:00:00+00:00           10681.20        10681.20             8   \n",
       "2008-01-14 21:00:00+00:00           12088.30        12088.30             9   \n",
       "2008-01-15 21:00:00+00:00           14302.40        14302.40            10   \n",
       "2008-01-16 21:00:00+00:00           15213.60        15213.60            11   \n",
       "2008-01-17 21:00:00+00:00           15964.00        15964.00            12   \n",
       "2008-01-18 21:00:00+00:00           17697.90        17697.90            13   \n",
       "2008-01-22 21:00:00+00:00           19363.20        19363.20            14   \n",
       "2008-01-23 21:00:00+00:00           20233.20        20233.20            15   \n",
       "2008-01-24 21:00:00+00:00           19470.08        19470.08            16   \n",
       "2008-01-25 21:00:00+00:00           20340.00        20340.00            17   \n",
       "2008-01-28 21:00:00+00:00           20801.44        20801.44            18   \n",
       "2008-01-29 21:00:00+00:00           22101.53        22101.53            19   \n",
       "2008-01-30 21:00:00+00:00           23677.20        23677.20            20   \n",
       "2008-01-31 21:00:00+00:00           25114.20        25114.20            21   \n",
       "2008-02-01 21:00:00+00:00           27072.00        27072.00            22   \n",
       "2008-02-04 21:00:00+00:00           28087.50        28087.50            23   \n",
       "2008-02-05 21:00:00+00:00           28963.00        28963.00            24   \n",
       "2008-02-06 21:00:00+00:00           29752.80        29752.80            25   \n",
       "2008-02-07 21:00:00+00:00           29280.00        29280.00            26   \n",
       "2008-02-08 21:00:00+00:00           30310.00        30310.00            27   \n",
       "2008-02-11 21:00:00+00:00           32624.80        32624.80            28   \n",
       "2008-02-12 21:00:00+00:00           34951.23        34951.23            29   \n",
       "2008-02-13 21:00:00+00:00           34960.80        34960.80            30   \n",
       "...                                      ...             ...           ...   \n",
       "2011-11-17 21:00:00+00:00         3763050.60      3763050.60           980   \n",
       "2011-11-18 21:00:00+00:00         3694843.90      3694843.90           981   \n",
       "2011-11-21 21:00:00+00:00         3674412.00      3674412.00           982   \n",
       "2011-11-22 21:00:00+00:00         3619988.10      3619988.10           983   \n",
       "2011-11-23 21:00:00+00:00         3697328.20      3697328.20           984   \n",
       "2011-11-25 18:00:00+00:00         3607511.70      3607511.70           985   \n",
       "2011-11-28 21:00:00+00:00         3577528.80      3577528.80           986   \n",
       "2011-11-29 21:00:00+00:00         3704782.00      3704782.00           987   \n",
       "2011-11-30 21:00:00+00:00         3679752.00      3679752.00           988   \n",
       "2011-12-01 21:00:00+00:00         3772314.00      3772314.00           989   \n",
       "2011-12-02 21:00:00+00:00         3832748.40      3832748.40           990   \n",
       "2011-12-05 21:00:00+00:00         3854133.00      3854133.00           991   \n",
       "2011-12-06 21:00:00+00:00         3890799.00      3890799.00           992   \n",
       "2011-12-07 21:00:00+00:00         3874314.50      3874314.50           993   \n",
       "2011-12-08 21:00:00+00:00         3859772.80      3859772.80           994   \n",
       "2011-12-09 21:00:00+00:00         3879253.80      3879253.80           995   \n",
       "2011-12-12 21:00:00+00:00         3912582.80      3912582.80           996   \n",
       "2011-12-13 21:00:00+00:00         3898808.00      3898808.00           997   \n",
       "2011-12-14 21:00:00+00:00         3872547.60      3872547.60           998   \n",
       "2011-12-15 21:00:00+00:00         3790494.30      3790494.30           999   \n",
       "2011-12-16 21:00:00+00:00         3781821.20      3781821.20          1000   \n",
       "2011-12-19 21:00:00+00:00         3806389.80      3806389.80          1001   \n",
       "2011-12-20 21:00:00+00:00         3822100.00      3822100.00          1002   \n",
       "2011-12-21 21:00:00+00:00         3963459.50      3963459.50          1003   \n",
       "2011-12-22 21:00:00+00:00         3972378.90      3972378.90          1004   \n",
       "2011-12-23 21:00:00+00:00         3997456.50      3997456.50          1005   \n",
       "2011-12-27 21:00:00+00:00         4049433.20      4049433.20          1006   \n",
       "2011-12-28 21:00:00+00:00         4085626.50      4085626.50          1007   \n",
       "2011-12-29 21:00:00+00:00         4050558.40      4050558.40          1008   \n",
       "2011-12-30 21:00:00+00:00         4079558.40      4079558.40          1009   \n",
       "\n",
       "                                                                transactions  \\\n",
       "2008-01-02 21:00:00+00:00                                                 []   \n",
       "2008-01-03 21:00:00+00:00  [{'price': 195.027465, 'order_id': '2802f8c018...   \n",
       "2008-01-04 21:00:00+00:00  [{'price': 180.140025, 'order_id': 'f603ddd817...   \n",
       "2008-01-07 21:00:00+00:00  [{'price': 177.72882, 'order_id': 'ffee7f5f1af...   \n",
       "2008-01-08 21:00:00+00:00  [{'price': 171.335625, 'order_id': '99be2796b4...   \n",
       "2008-01-09 21:00:00+00:00  [{'price': 179.4897, 'order_id': '26a631854700...   \n",
       "2008-01-10 21:00:00+00:00  [{'price': 178.10901, 'order_id': 'b5c4b795bbf...   \n",
       "2008-01-11 21:00:00+00:00  [{'price': 172.776345, 'order_id': '83ba09ccc5...   \n",
       "2008-01-14 21:00:00+00:00  [{'price': 178.86939, 'order_id': '783bf3e154c...   \n",
       "2008-01-15 21:00:00+00:00  [{'price': 169.12452, 'order_id': '7603a588085...   \n",
       "2008-01-16 21:00:00+00:00  [{'price': 159.71982000000003, 'order_id': 'e4...   \n",
       "2008-01-17 21:00:00+00:00  [{'price': 160.970445, 'order_id': '11794fdf10...   \n",
       "2008-01-18 21:00:00+00:00  [{'price': 161.44068000000001, 'order_id': '40...   \n",
       "2008-01-22 21:00:00+00:00  [{'price': 155.71782000000002, 'order_id': 'ba...   \n",
       "2008-01-23 21:00:00+00:00  [{'price': 139.141536, 'order_id': 'a9c98d339f...   \n",
       "2008-01-24 21:00:00+00:00  [{'price': 135.6678, 'order_id': '70d38e218e49...   \n",
       "2008-01-25 21:00:00+00:00  [{'price': 130.0740045, 'order_id': '01259c426...   \n",
       "2008-01-28 21:00:00+00:00  [{'price': 130.0740045, 'order_id': 'dd5acc73d...   \n",
       "2008-01-29 21:00:00+00:00  [{'price': 131.60576999999998, 'order_id': '97...   \n",
       "2008-01-30 21:00:00+00:00  [{'price': 132.24609, 'order_id': '3fd3dc2a462...   \n",
       "2008-01-31 21:00:00+00:00  [{'price': 135.42768, 'order_id': 'b8b32c32b11...   \n",
       "2008-02-01 21:00:00+00:00  [{'price': 133.816875, 'order_id': 'd33536fcc2...   \n",
       "2008-02-04 21:00:00+00:00  [{'price': 131.715825, 'order_id': '7be76df3f4...   \n",
       "2008-02-05 21:00:00+00:00  [{'price': 129.42468000000002, 'order_id': 'ab...   \n",
       "2008-02-06 21:00:00+00:00  [{'price': 122.061, 'order_id': 'a934f3a3e4724...   \n",
       "2008-02-07 21:00:00+00:00  [{'price': 121.30062000000001, 'order_id': '2f...   \n",
       "2008-02-08 21:00:00+00:00  [{'price': 125.54274000000001, 'order_id': 'ab...   \n",
       "2008-02-11 21:00:00+00:00  [{'price': 129.51372450000002, 'order_id': '2d...   \n",
       "2008-02-12 21:00:00+00:00  [{'price': 124.92243, 'order_id': 'b415841c42f...   \n",
       "2008-02-13 21:00:00+00:00  [{'price': 129.4647, 'order_id': 'c4c8109ced65...   \n",
       "...                                                                      ...   \n",
       "2011-11-17 21:00:00+00:00  [{'price': 377.59870500000005, 'order_id': '6a...   \n",
       "2011-11-18 21:00:00+00:00  [{'price': 375.12747, 'order_id': 'a78d598a213...   \n",
       "2011-11-21 21:00:00+00:00  [{'price': 369.194505, 'order_id': '14ef45fb2d...   \n",
       "2011-11-22 21:00:00+00:00  [{'price': 376.698255, 'order_id': '1add0085de...   \n",
       "2011-11-23 21:00:00+00:00  [{'price': 367.173495, 'order_id': 'd9d979320e...   \n",
       "2011-11-25 18:00:00+00:00  [{'price': 363.751785, 'order_id': '94e4a4dc0f...   \n",
       "2011-11-28 21:00:00+00:00  [{'price': 376.30806, 'order_id': '26d5e56042d...   \n",
       "2011-11-29 21:00:00+00:00  [{'price': 373.3866, 'order_id': 'dd7d39c74332...   \n",
       "2011-11-30 21:00:00+00:00  [{'price': 382.3911, 'order_id': 'c8d47ba83a0f...   \n",
       "2011-12-01 21:00:00+00:00  [{'price': 388.123965, 'order_id': '4103ec0d23...   \n",
       "2011-12-02 21:00:00+00:00  [{'price': 389.89484999999996, 'order_id': 'ef...   \n",
       "2011-12-05 21:00:00+00:00  [{'price': 393.206505, 'order_id': 'b4ac822284...   \n",
       "2011-12-06 21:00:00+00:00  [{'price': 391.145475, 'order_id': '487c2534a2...   \n",
       "2011-12-07 21:00:00+00:00  [{'price': 389.28454500000004, 'order_id': 'ed...   \n",
       "2011-12-08 21:00:00+00:00  [{'price': 390.85533000000004, 'order_id': 'bb...   \n",
       "2011-12-09 21:00:00+00:00  [{'price': 393.81681000000003, 'order_id': '6e...   \n",
       "2011-12-12 21:00:00+00:00  [{'price': 392.03592000000003, 'order_id': 'dd...   \n",
       "2011-12-13 21:00:00+00:00  [{'price': 389.004405, 'order_id': '8d03848e27...   \n",
       "2011-12-14 21:00:00+00:00  [{'price': 380.380095, 'order_id': '9ed89feec7...   \n",
       "2011-12-15 21:00:00+00:00  [{'price': 379.12946999999997, 'order_id': 'b1...   \n",
       "2011-12-16 21:00:00+00:00  [{'price': 381.21051, 'order_id': 'd57b785ed86...   \n",
       "2011-12-19 21:00:00+00:00  [{'price': 382.401105, 'order_id': '6a1cf5b26a...   \n",
       "2011-12-20 21:00:00+00:00  [{'price': 396.147975, 'order_id': 'c42d08b8be...   \n",
       "2011-12-21 21:00:00+00:00  [{'price': 396.6432225, 'order_id': '94695baad...   \n",
       "2011-12-22 21:00:00+00:00  [{'price': 398.749275, 'order_id': '92debfff54...   \n",
       "2011-12-23 21:00:00+00:00  [{'price': 403.531665, 'order_id': '30c49c6213...   \n",
       "2011-12-27 21:00:00+00:00  [{'price': 406.733265, 'order_id': '3f153af629...   \n",
       "2011-12-28 21:00:00+00:00  [{'price': 402.84132, 'order_id': '58ed60a8ebc...   \n",
       "2011-12-29 21:00:00+00:00  [{'price': 405.32256, 'order_id': '21fa35f5323...   \n",
       "2011-12-30 21:00:00+00:00  [{'price': 405.2025, 'order_id': 'c47392309589...   \n",
       "\n",
       "                          treasury_period_return  \n",
       "2008-01-02 21:00:00+00:00                    0.0  \n",
       "2008-01-03 21:00:00+00:00                    0.0  \n",
       "2008-01-04 21:00:00+00:00                    0.0  \n",
       "2008-01-07 21:00:00+00:00                    0.0  \n",
       "2008-01-08 21:00:00+00:00                    0.0  \n",
       "2008-01-09 21:00:00+00:00                    0.0  \n",
       "2008-01-10 21:00:00+00:00                    0.0  \n",
       "2008-01-11 21:00:00+00:00                    0.0  \n",
       "2008-01-14 21:00:00+00:00                    0.0  \n",
       "2008-01-15 21:00:00+00:00                    0.0  \n",
       "2008-01-16 21:00:00+00:00                    0.0  \n",
       "2008-01-17 21:00:00+00:00                    0.0  \n",
       "2008-01-18 21:00:00+00:00                    0.0  \n",
       "2008-01-22 21:00:00+00:00                    0.0  \n",
       "2008-01-23 21:00:00+00:00                    0.0  \n",
       "2008-01-24 21:00:00+00:00                    0.0  \n",
       "2008-01-25 21:00:00+00:00                    0.0  \n",
       "2008-01-28 21:00:00+00:00                    0.0  \n",
       "2008-01-29 21:00:00+00:00                    0.0  \n",
       "2008-01-30 21:00:00+00:00                    0.0  \n",
       "2008-01-31 21:00:00+00:00                    0.0  \n",
       "2008-02-01 21:00:00+00:00                    0.0  \n",
       "2008-02-04 21:00:00+00:00                    0.0  \n",
       "2008-02-05 21:00:00+00:00                    0.0  \n",
       "2008-02-06 21:00:00+00:00                    0.0  \n",
       "2008-02-07 21:00:00+00:00                    0.0  \n",
       "2008-02-08 21:00:00+00:00                    0.0  \n",
       "2008-02-11 21:00:00+00:00                    0.0  \n",
       "2008-02-12 21:00:00+00:00                    0.0  \n",
       "2008-02-13 21:00:00+00:00                    0.0  \n",
       "...                                          ...  \n",
       "2011-11-17 21:00:00+00:00                    0.0  \n",
       "2011-11-18 21:00:00+00:00                    0.0  \n",
       "2011-11-21 21:00:00+00:00                    0.0  \n",
       "2011-11-22 21:00:00+00:00                    0.0  \n",
       "2011-11-23 21:00:00+00:00                    0.0  \n",
       "2011-11-25 18:00:00+00:00                    0.0  \n",
       "2011-11-28 21:00:00+00:00                    0.0  \n",
       "2011-11-29 21:00:00+00:00                    0.0  \n",
       "2011-11-30 21:00:00+00:00                    0.0  \n",
       "2011-12-01 21:00:00+00:00                    0.0  \n",
       "2011-12-02 21:00:00+00:00                    0.0  \n",
       "2011-12-05 21:00:00+00:00                    0.0  \n",
       "2011-12-06 21:00:00+00:00                    0.0  \n",
       "2011-12-07 21:00:00+00:00                    0.0  \n",
       "2011-12-08 21:00:00+00:00                    0.0  \n",
       "2011-12-09 21:00:00+00:00                    0.0  \n",
       "2011-12-12 21:00:00+00:00                    0.0  \n",
       "2011-12-13 21:00:00+00:00                    0.0  \n",
       "2011-12-14 21:00:00+00:00                    0.0  \n",
       "2011-12-15 21:00:00+00:00                    0.0  \n",
       "2011-12-16 21:00:00+00:00                    0.0  \n",
       "2011-12-19 21:00:00+00:00                    0.0  \n",
       "2011-12-20 21:00:00+00:00                    0.0  \n",
       "2011-12-21 21:00:00+00:00                    0.0  \n",
       "2011-12-22 21:00:00+00:00                    0.0  \n",
       "2011-12-23 21:00:00+00:00                    0.0  \n",
       "2011-12-27 21:00:00+00:00                    0.0  \n",
       "2011-12-28 21:00:00+00:00                    0.0  \n",
       "2011-12-29 21:00:00+00:00                    0.0  \n",
       "2011-12-30 21:00:00+00:00                    0.0  \n",
       "\n",
       "[1009 rows x 38 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Collect Apple's share price data from 2008 to 2012, using Zipline\n",
    "%zipline --bundle quantopian-quandl --start 2008-1-1 --end 2012-1-1 -o strat.pickle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"visstrat\"></a> Visualizing Strategy Metrics\n",
    "\n",
    "Having saved our strategy output to ```strat.pickle```, above, we will here read from that saved data file and visualize it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "\n",
    "style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "backtest_df = pd.read_pickle('strat.pickle')\n",
    "backtest_df.portfolio_value.plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The kinds of data saved in our backtest data file can be observed in its columns:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['AAPL', 'algo_volatility', 'algorithm_period_return', 'alpha',\n",
       "       'benchmark_period_return', 'benchmark_volatility', 'beta',\n",
       "       'capital_used', 'ending_cash', 'ending_exposure', 'ending_value',\n",
       "       'excess_return', 'gross_leverage', 'long_exposure', 'long_value',\n",
       "       'longs_count', 'max_drawdown', 'max_leverage', 'net_leverage', 'orders',\n",
       "       'period_close', 'period_label', 'period_open', 'pnl', 'portfolio_value',\n",
       "       'positions', 'returns', 'sharpe', 'short_exposure', 'short_value',\n",
       "       'shorts_count', 'sortino', 'starting_cash', 'starting_exposure',\n",
       "       'starting_value', 'trading_days', 'transactions',\n",
       "       'treasury_period_return'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "backtest_df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This process can be made more streamlined, by including the following ```analyze``` function. Anything included in this function will run automatically at the end of the jupyter ```magic``` function is executed below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def handle_data(context, data):\n",
    "    order(symbol('AAPL'), 10)\n",
    "    record(AAPL=data.current(symbol('AAPL'), 'price'))\n",
    "\n",
    "def analyze(context, perf):\n",
    "    fig = plt.figure()\n",
    "    ax1 = fig.add_subplot(111)\n",
    "    perf.portfolio_value.plot(ax=ax1)\n",
    "    ax1.set_ylabel('portfolio & value')\n",
    "    plt.legend(loc=0)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>algo_volatility</th>\n",
       "      <th>algorithm_period_return</th>\n",
       "      <th>alpha</th>\n",
       "      <th>benchmark_period_return</th>\n",
       "      <th>benchmark_volatility</th>\n",
       "      <th>beta</th>\n",
       "      <th>capital_used</th>\n",
       "      <th>ending_cash</th>\n",
       "      <th>ending_exposure</th>\n",
       "      <th>...</th>\n",
       "      <th>short_exposure</th>\n",
       "      <th>short_value</th>\n",
       "      <th>shorts_count</th>\n",
       "      <th>sortino</th>\n",
       "      <th>starting_cash</th>\n",
       "      <th>starting_exposure</th>\n",
       "      <th>starting_value</th>\n",
       "      <th>trading_days</th>\n",
       "      <th>transactions</th>\n",
       "      <th>treasury_period_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2008-01-02 21:00:00+00:00</th>\n",
       "      <td>194.840</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.003665</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1</td>\n",
       "      <td>[]</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-03 21:00:00+00:00</th>\n",
       "      <td>194.930</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>-9.846500e-08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.007343</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1950.284650</td>\n",
       "      <td>9.998050e+06</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-11.224972</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2</td>\n",
       "      <td>[{'price': 195.027465, 'order_id': 'ec52ad7022...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-04 21:00:00+00:00</th>\n",
       "      <td>180.050</td>\n",
       "      <td>0.000137</td>\n",
       "      <td>-1.506949e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.011035</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1801.410250</td>\n",
       "      <td>9.996248e+06</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-9.225231</td>\n",
       "      <td>9.998050e+06</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>1949.30</td>\n",
       "      <td>3</td>\n",
       "      <td>[{'price': 180.140025, 'order_id': '6448608787...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-07 21:00:00+00:00</th>\n",
       "      <td>177.640</td>\n",
       "      <td>0.000112</td>\n",
       "      <td>-1.997931e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.014741</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1777.298200</td>\n",
       "      <td>9.994471e+06</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-10.064893</td>\n",
       "      <td>9.996248e+06</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>3601.00</td>\n",
       "      <td>4</td>\n",
       "      <td>[{'price': 177.72882, 'order_id': '24bc1361705...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-08 21:00:00+00:00</th>\n",
       "      <td>171.250</td>\n",
       "      <td>0.000140</td>\n",
       "      <td>-3.923593e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.018460</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1713.366250</td>\n",
       "      <td>9.992758e+06</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-11.195170</td>\n",
       "      <td>9.994471e+06</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>5329.20</td>\n",
       "      <td>5</td>\n",
       "      <td>[{'price': 171.335625, 'order_id': 'cd411aad25...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-09 21:00:00+00:00</th>\n",
       "      <td>179.400</td>\n",
       "      <td>0.000290</td>\n",
       "      <td>-6.726635e-06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.022192</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1794.907000</td>\n",
       "      <td>9.990963e+06</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.751847</td>\n",
       "      <td>9.992758e+06</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>6850.00</td>\n",
       "      <td>6</td>\n",
       "      <td>[{'price': 179.4897, 'order_id': '2b3a53bd317c...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-10 21:00:00+00:00</th>\n",
       "      <td>178.020</td>\n",
       "      <td>0.000267</td>\n",
       "      <td>-1.371664e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.025939</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1781.100100</td>\n",
       "      <td>9.989182e+06</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.184238</td>\n",
       "      <td>9.990963e+06</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>8970.00</td>\n",
       "      <td>7</td>\n",
       "      <td>[{'price': 178.10901, 'order_id': '4e25a9996e6...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-11 21:00:00+00:00</th>\n",
       "      <td>172.690</td>\n",
       "      <td>0.000299</td>\n",
       "      <td>-4.578399e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.029699</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1727.773450</td>\n",
       "      <td>9.987454e+06</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.239021</td>\n",
       "      <td>9.989182e+06</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>10681.20</td>\n",
       "      <td>8</td>\n",
       "      <td>[{'price': 172.776345, 'order_id': '39f18b2e27...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-14 21:00:00+00:00</th>\n",
       "      <td>178.780</td>\n",
       "      <td>0.000379</td>\n",
       "      <td>-3.244380e-06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033473</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1788.703900</td>\n",
       "      <td>9.985665e+06</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.416540</td>\n",
       "      <td>9.987454e+06</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>12088.30</td>\n",
       "      <td>9</td>\n",
       "      <td>[{'price': 178.86939, 'order_id': 'dc8a32a5821...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-15 21:00:00+00:00</th>\n",
       "      <td>169.040</td>\n",
       "      <td>0.000529</td>\n",
       "      <td>-8.124990e-05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.037260</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1691.255200</td>\n",
       "      <td>9.983974e+06</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-4.623698</td>\n",
       "      <td>9.985665e+06</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>14302.40</td>\n",
       "      <td>10</td>\n",
       "      <td>[{'price': 169.12452, 'order_id': '2fc850446ed...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-16 21:00:00+00:00</th>\n",
       "      <td>159.640</td>\n",
       "      <td>0.000621</td>\n",
       "      <td>-1.659307e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.041062</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1597.208200</td>\n",
       "      <td>9.982377e+06</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.494960</td>\n",
       "      <td>9.983974e+06</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>15213.60</td>\n",
       "      <td>11</td>\n",
       "      <td>[{'price': 159.71982000000003, 'order_id': '7e...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-17 21:00:00+00:00</th>\n",
       "      <td>160.890</td>\n",
       "      <td>0.000606</td>\n",
       "      <td>-1.535122e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.044877</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1609.714450</td>\n",
       "      <td>9.980767e+06</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.752987</td>\n",
       "      <td>9.982377e+06</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>15964.00</td>\n",
       "      <td>12</td>\n",
       "      <td>[{'price': 160.970445, 'order_id': 'bb06e2249a...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-18 21:00:00+00:00</th>\n",
       "      <td>161.360</td>\n",
       "      <td>0.000585</td>\n",
       "      <td>-1.484238e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.048707</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1614.416800</td>\n",
       "      <td>9.979153e+06</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.344059</td>\n",
       "      <td>9.980767e+06</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>17697.90</td>\n",
       "      <td>13</td>\n",
       "      <td>[{'price': 161.44068000000001, 'order_id': 'df...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-22 21:00:00+00:00</th>\n",
       "      <td>155.640</td>\n",
       "      <td>0.000613</td>\n",
       "      <td>-2.171427e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.052550</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1557.188200</td>\n",
       "      <td>9.977595e+06</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.567920</td>\n",
       "      <td>9.979153e+06</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>19363.20</td>\n",
       "      <td>14</td>\n",
       "      <td>[{'price': 155.71782000000002, 'order_id': '3a...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-23 21:00:00+00:00</th>\n",
       "      <td>139.072</td>\n",
       "      <td>0.001010</td>\n",
       "      <td>-4.325972e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.056408</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1391.425360</td>\n",
       "      <td>9.976204e+06</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.896659</td>\n",
       "      <td>9.977595e+06</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>20233.20</td>\n",
       "      <td>15</td>\n",
       "      <td>[{'price': 139.141536, 'order_id': '4dc593488f...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-24 21:00:00+00:00</th>\n",
       "      <td>135.600</td>\n",
       "      <td>0.000979</td>\n",
       "      <td>-4.812740e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.060280</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1356.688000</td>\n",
       "      <td>9.974847e+06</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.299514</td>\n",
       "      <td>9.976204e+06</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>19470.08</td>\n",
       "      <td>16</td>\n",
       "      <td>[{'price': 135.6678, 'order_id': '95328d0e66bc...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-25 21:00:00+00:00</th>\n",
       "      <td>130.009</td>\n",
       "      <td>0.000970</td>\n",
       "      <td>-5.652050e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.064166</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1300.750045</td>\n",
       "      <td>9.973547e+06</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.919264</td>\n",
       "      <td>9.974847e+06</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>20340.00</td>\n",
       "      <td>17</td>\n",
       "      <td>[{'price': 130.0740045, 'order_id': 'aff646a2c...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-28 21:00:00+00:00</th>\n",
       "      <td>130.009</td>\n",
       "      <td>0.000950</td>\n",
       "      <td>-5.652710e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.068066</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1300.750045</td>\n",
       "      <td>9.972246e+06</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.697040</td>\n",
       "      <td>9.973547e+06</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>20801.44</td>\n",
       "      <td>18</td>\n",
       "      <td>[{'price': 130.0740045, 'order_id': '983ef46c1...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-29 21:00:00+00:00</th>\n",
       "      <td>131.540</td>\n",
       "      <td>0.000946</td>\n",
       "      <td>-5.393108e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.071980</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1316.067700</td>\n",
       "      <td>9.970930e+06</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.147567</td>\n",
       "      <td>9.972246e+06</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>22101.53</td>\n",
       "      <td>19</td>\n",
       "      <td>[{'price': 131.60576999999998, 'order_id': 'a6...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-30 21:00:00+00:00</th>\n",
       "      <td>132.180</td>\n",
       "      <td>0.000932</td>\n",
       "      <td>-5.278579e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.075909</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1322.470900</td>\n",
       "      <td>9.969607e+06</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.818592</td>\n",
       "      <td>9.970930e+06</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>23677.20</td>\n",
       "      <td>20</td>\n",
       "      <td>[{'price': 132.24609, 'order_id': '3f20771de51...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-31 21:00:00+00:00</th>\n",
       "      <td>135.360</td>\n",
       "      <td>0.000957</td>\n",
       "      <td>-4.675065e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.079852</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1354.286800</td>\n",
       "      <td>9.968253e+06</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.893207</td>\n",
       "      <td>9.969607e+06</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>25114.20</td>\n",
       "      <td>21</td>\n",
       "      <td>[{'price': 135.42768, 'order_id': '372d6c56fd5...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-01 21:00:00+00:00</th>\n",
       "      <td>133.750</td>\n",
       "      <td>0.000934</td>\n",
       "      <td>-4.997744e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.083810</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1338.178750</td>\n",
       "      <td>9.966915e+06</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.113220</td>\n",
       "      <td>9.968253e+06</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>27072.00</td>\n",
       "      <td>22</td>\n",
       "      <td>[{'price': 133.816875, 'order_id': '5a9d61f73b...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-04 21:00:00+00:00</th>\n",
       "      <td>131.650</td>\n",
       "      <td>0.000915</td>\n",
       "      <td>-5.439412e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.087782</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1317.168250</td>\n",
       "      <td>9.965598e+06</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.425988</td>\n",
       "      <td>9.966915e+06</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>28087.50</td>\n",
       "      <td>23</td>\n",
       "      <td>[{'price': 131.715825, 'order_id': '3ff3b04eaf...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-05 21:00:00+00:00</th>\n",
       "      <td>129.360</td>\n",
       "      <td>0.000900</td>\n",
       "      <td>-5.943869e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.091769</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1294.256800</td>\n",
       "      <td>9.964303e+06</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.765445</td>\n",
       "      <td>9.965598e+06</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>28963.00</td>\n",
       "      <td>24</td>\n",
       "      <td>[{'price': 129.42468000000002, 'order_id': '25...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-06 21:00:00+00:00</th>\n",
       "      <td>122.000</td>\n",
       "      <td>0.000993</td>\n",
       "      <td>-7.637289e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.095770</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1220.620000</td>\n",
       "      <td>9.963083e+06</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.320042</td>\n",
       "      <td>9.964303e+06</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>29752.80</td>\n",
       "      <td>25</td>\n",
       "      <td>[{'price': 122.061, 'order_id': '1bc784b04e2b4...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-07 21:00:00+00:00</th>\n",
       "      <td>121.240</td>\n",
       "      <td>0.000974</td>\n",
       "      <td>-7.820305e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.099786</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1213.016200</td>\n",
       "      <td>9.961870e+06</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.338775</td>\n",
       "      <td>9.963083e+06</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>29280.00</td>\n",
       "      <td>26</td>\n",
       "      <td>[{'price': 121.30062000000001, 'order_id': 'c6...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-08 21:00:00+00:00</th>\n",
       "      <td>125.480</td>\n",
       "      <td>0.001042</td>\n",
       "      <td>-6.760943e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.103817</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1255.437400</td>\n",
       "      <td>9.960614e+06</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.225585</td>\n",
       "      <td>9.961870e+06</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>30310.00</td>\n",
       "      <td>27</td>\n",
       "      <td>[{'price': 125.54274000000001, 'order_id': 'b7...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-11 21:00:00+00:00</th>\n",
       "      <td>129.449</td>\n",
       "      <td>0.001092</td>\n",
       "      <td>-5.729660e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.107862</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1295.147245</td>\n",
       "      <td>9.959319e+06</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.180489</td>\n",
       "      <td>9.960614e+06</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>32624.80</td>\n",
       "      <td>28</td>\n",
       "      <td>[{'price': 129.51372450000002, 'order_id': '92...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-12 21:00:00+00:00</th>\n",
       "      <td>124.860</td>\n",
       "      <td>0.001115</td>\n",
       "      <td>-6.969324e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.111922</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1249.234300</td>\n",
       "      <td>9.958070e+06</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-5.800272</td>\n",
       "      <td>9.959319e+06</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>34951.23</td>\n",
       "      <td>29</td>\n",
       "      <td>[{'price': 124.92243, 'order_id': '194f18a4306...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-13 21:00:00+00:00</th>\n",
       "      <td>129.400</td>\n",
       "      <td>0.001180</td>\n",
       "      <td>-5.698781e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.115998</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1294.657000</td>\n",
       "      <td>9.956775e+06</td>\n",
       "      <td>37526.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-4.662658</td>\n",
       "      <td>9.958070e+06</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>34960.80</td>\n",
       "      <td>30</td>\n",
       "      <td>[{'price': 129.4647, 'order_id': '7a1b6925d46d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-17 21:00:00+00:00</th>\n",
       "      <td>377.410</td>\n",
       "      <td>0.042831</td>\n",
       "      <td>1.506754e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.017863</td>\n",
       "      <td>0.005238</td>\n",
       "      <td>-0.575354</td>\n",
       "      <td>-3775.997050</td>\n",
       "      <td>7.811911e+06</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.265484</td>\n",
       "      <td>7.815687e+06</td>\n",
       "      <td>3763050.60</td>\n",
       "      <td>3763050.60</td>\n",
       "      <td>980</td>\n",
       "      <td>[{'price': 377.59870500000005, 'order_id': '8c...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-18 21:00:00+00:00</th>\n",
       "      <td>374.940</td>\n",
       "      <td>0.042825</td>\n",
       "      <td>1.482571e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.146203</td>\n",
       "      <td>0.005235</td>\n",
       "      <td>-0.575979</td>\n",
       "      <td>-3751.284700</td>\n",
       "      <td>7.808159e+06</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.245542</td>\n",
       "      <td>7.811911e+06</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>3694843.90</td>\n",
       "      <td>981</td>\n",
       "      <td>[{'price': 375.12747, 'order_id': '44182aeb52b...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-21 21:00:00+00:00</th>\n",
       "      <td>369.010</td>\n",
       "      <td>0.042884</td>\n",
       "      <td>1.424455e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.275013</td>\n",
       "      <td>0.005232</td>\n",
       "      <td>-0.577433</td>\n",
       "      <td>-3691.955050</td>\n",
       "      <td>7.804467e+06</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.195890</td>\n",
       "      <td>7.808159e+06</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>3674412.00</td>\n",
       "      <td>982</td>\n",
       "      <td>[{'price': 369.194505, 'order_id': 'a41897b256...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-22 21:00:00+00:00</th>\n",
       "      <td>376.510</td>\n",
       "      <td>0.042981</td>\n",
       "      <td>1.498029e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.404296</td>\n",
       "      <td>0.005230</td>\n",
       "      <td>-0.575665</td>\n",
       "      <td>-3766.992550</td>\n",
       "      <td>7.800700e+06</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.251567</td>\n",
       "      <td>7.804467e+06</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>3619988.10</td>\n",
       "      <td>983</td>\n",
       "      <td>[{'price': 376.698255, 'order_id': 'c4ad06aee8...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-23 21:00:00+00:00</th>\n",
       "      <td>366.990</td>\n",
       "      <td>0.043163</td>\n",
       "      <td>1.404540e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.534052</td>\n",
       "      <td>0.005227</td>\n",
       "      <td>-0.577974</td>\n",
       "      <td>-3671.744950</td>\n",
       "      <td>7.797029e+06</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.168451</td>\n",
       "      <td>7.800700e+06</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>3697328.20</td>\n",
       "      <td>984</td>\n",
       "      <td>[{'price': 367.173495, 'order_id': '108fae2a48...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-25 18:00:00+00:00</th>\n",
       "      <td>363.570</td>\n",
       "      <td>0.043169</td>\n",
       "      <td>1.370920e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.664284</td>\n",
       "      <td>0.005225</td>\n",
       "      <td>-0.578831</td>\n",
       "      <td>-3637.527850</td>\n",
       "      <td>7.793391e+06</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.140921</td>\n",
       "      <td>7.797029e+06</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>3607511.70</td>\n",
       "      <td>985</td>\n",
       "      <td>[{'price': 363.751785, 'order_id': '50af5af4c1...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-28 21:00:00+00:00</th>\n",
       "      <td>376.120</td>\n",
       "      <td>0.043486</td>\n",
       "      <td>1.494410e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.794994</td>\n",
       "      <td>0.005222</td>\n",
       "      <td>-0.575833</td>\n",
       "      <td>-3763.090600</td>\n",
       "      <td>7.789628e+06</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.234077</td>\n",
       "      <td>7.793391e+06</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>3577528.80</td>\n",
       "      <td>986</td>\n",
       "      <td>[{'price': 376.30806, 'order_id': '7e9422a0925...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-29 21:00:00+00:00</th>\n",
       "      <td>373.200</td>\n",
       "      <td>0.043485</td>\n",
       "      <td>1.465646e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>34.926182</td>\n",
       "      <td>0.005219</td>\n",
       "      <td>-0.576566</td>\n",
       "      <td>-3733.876000</td>\n",
       "      <td>7.785894e+06</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.210759</td>\n",
       "      <td>7.789628e+06</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>3704782.00</td>\n",
       "      <td>987</td>\n",
       "      <td>[{'price': 373.3866, 'order_id': '7be7f3afe333...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-11-30 21:00:00+00:00</th>\n",
       "      <td>382.200</td>\n",
       "      <td>0.043632</td>\n",
       "      <td>1.554384e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.057851</td>\n",
       "      <td>0.005217</td>\n",
       "      <td>-0.574446</td>\n",
       "      <td>-3823.921000</td>\n",
       "      <td>7.782070e+06</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.276818</td>\n",
       "      <td>7.785894e+06</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>3679752.00</td>\n",
       "      <td>988</td>\n",
       "      <td>[{'price': 382.3911, 'order_id': '8257dea0cf46...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-01 21:00:00+00:00</th>\n",
       "      <td>387.930</td>\n",
       "      <td>0.043675</td>\n",
       "      <td>1.610937e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.190003</td>\n",
       "      <td>0.005214</td>\n",
       "      <td>-0.573122</td>\n",
       "      <td>-3881.249650</td>\n",
       "      <td>7.778189e+06</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.318314</td>\n",
       "      <td>7.782070e+06</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>3772314.00</td>\n",
       "      <td>989</td>\n",
       "      <td>[{'price': 388.123965, 'order_id': '410334c727...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-02 21:00:00+00:00</th>\n",
       "      <td>389.700</td>\n",
       "      <td>0.043659</td>\n",
       "      <td>1.628423e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.322639</td>\n",
       "      <td>0.005211</td>\n",
       "      <td>-0.572742</td>\n",
       "      <td>-3898.958500</td>\n",
       "      <td>7.774290e+06</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.330608</td>\n",
       "      <td>7.778189e+06</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>3832748.40</td>\n",
       "      <td>990</td>\n",
       "      <td>[{'price': 389.89484999999996, 'order_id': 'ad...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-05 21:00:00+00:00</th>\n",
       "      <td>393.010</td>\n",
       "      <td>0.043657</td>\n",
       "      <td>1.661157e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.455761</td>\n",
       "      <td>0.005209</td>\n",
       "      <td>-0.571999</td>\n",
       "      <td>-3932.075050</td>\n",
       "      <td>7.770358e+06</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.354150</td>\n",
       "      <td>7.774290e+06</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>3854133.00</td>\n",
       "      <td>991</td>\n",
       "      <td>[{'price': 393.206505, 'order_id': 'cdb699ea05...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-06 21:00:00+00:00</th>\n",
       "      <td>390.950</td>\n",
       "      <td>0.043646</td>\n",
       "      <td>1.640761e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.589371</td>\n",
       "      <td>0.005206</td>\n",
       "      <td>-0.572524</td>\n",
       "      <td>-3911.464750</td>\n",
       "      <td>7.766446e+06</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.337835</td>\n",
       "      <td>7.770358e+06</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>3890799.00</td>\n",
       "      <td>992</td>\n",
       "      <td>[{'price': 391.145475, 'order_id': '8686544864...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-07 21:00:00+00:00</th>\n",
       "      <td>389.090</td>\n",
       "      <td>0.043633</td>\n",
       "      <td>1.622326e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.723471</td>\n",
       "      <td>0.005204</td>\n",
       "      <td>-0.573001</td>\n",
       "      <td>-3892.855450</td>\n",
       "      <td>7.762554e+06</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.323077</td>\n",
       "      <td>7.766446e+06</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>3874314.50</td>\n",
       "      <td>993</td>\n",
       "      <td>[{'price': 389.28454500000004, 'order_id': '3b...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-08 21:00:00+00:00</th>\n",
       "      <td>390.660</td>\n",
       "      <td>0.043615</td>\n",
       "      <td>1.637899e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.858062</td>\n",
       "      <td>0.005201</td>\n",
       "      <td>-0.572668</td>\n",
       "      <td>-3908.563300</td>\n",
       "      <td>7.758645e+06</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.333910</td>\n",
       "      <td>7.762554e+06</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>3859772.80</td>\n",
       "      <td>994</td>\n",
       "      <td>[{'price': 390.85533000000004, 'order_id': '78...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-09 21:00:00+00:00</th>\n",
       "      <td>393.620</td>\n",
       "      <td>0.043609</td>\n",
       "      <td>1.667290e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>35.993146</td>\n",
       "      <td>0.005198</td>\n",
       "      <td>-0.572009</td>\n",
       "      <td>-3938.178100</td>\n",
       "      <td>7.754707e+06</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.354900</td>\n",
       "      <td>7.758645e+06</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>3879253.80</td>\n",
       "      <td>995</td>\n",
       "      <td>[{'price': 393.81681000000003, 'order_id': 'c0...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-12 21:00:00+00:00</th>\n",
       "      <td>391.840</td>\n",
       "      <td>0.043595</td>\n",
       "      <td>1.649594e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.128726</td>\n",
       "      <td>0.005196</td>\n",
       "      <td>-0.572467</td>\n",
       "      <td>-3920.369200</td>\n",
       "      <td>7.750786e+06</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.340770</td>\n",
       "      <td>7.754707e+06</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>3912582.80</td>\n",
       "      <td>996</td>\n",
       "      <td>[{'price': 392.03592000000003, 'order_id': '39...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-13 21:00:00+00:00</th>\n",
       "      <td>388.810</td>\n",
       "      <td>0.043595</td>\n",
       "      <td>1.619444e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.264802</td>\n",
       "      <td>0.005193</td>\n",
       "      <td>-0.573220</td>\n",
       "      <td>-3890.054050</td>\n",
       "      <td>7.746896e+06</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.316649</td>\n",
       "      <td>7.750786e+06</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>3898808.00</td>\n",
       "      <td>997</td>\n",
       "      <td>[{'price': 389.004405, 'order_id': '5fa77db006...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-14 21:00:00+00:00</th>\n",
       "      <td>380.190</td>\n",
       "      <td>0.043738</td>\n",
       "      <td>1.533587e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.401377</td>\n",
       "      <td>0.005191</td>\n",
       "      <td>-0.575294</td>\n",
       "      <td>-3803.810950</td>\n",
       "      <td>7.743093e+06</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.242998</td>\n",
       "      <td>7.746896e+06</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>3872547.60</td>\n",
       "      <td>998</td>\n",
       "      <td>[{'price': 380.380095, 'order_id': '8ce63b39bf...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-15 21:00:00+00:00</th>\n",
       "      <td>378.940</td>\n",
       "      <td>0.043720</td>\n",
       "      <td>1.521122e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.538453</td>\n",
       "      <td>0.005188</td>\n",
       "      <td>-0.575627</td>\n",
       "      <td>-3791.304700</td>\n",
       "      <td>7.739301e+06</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.233004</td>\n",
       "      <td>7.743093e+06</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>3790494.30</td>\n",
       "      <td>999</td>\n",
       "      <td>[{'price': 379.12946999999997, 'order_id': '67...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-16 21:00:00+00:00</th>\n",
       "      <td>381.020</td>\n",
       "      <td>0.043706</td>\n",
       "      <td>1.541879e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.676031</td>\n",
       "      <td>0.005186</td>\n",
       "      <td>-0.575167</td>\n",
       "      <td>-3812.115100</td>\n",
       "      <td>7.735489e+06</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.247658</td>\n",
       "      <td>7.739301e+06</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>3781821.20</td>\n",
       "      <td>1000</td>\n",
       "      <td>[{'price': 381.21051, 'order_id': 'a814483b0ed...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-19 21:00:00+00:00</th>\n",
       "      <td>382.210</td>\n",
       "      <td>0.043687</td>\n",
       "      <td>1.553765e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.814113</td>\n",
       "      <td>0.005183</td>\n",
       "      <td>-0.574920</td>\n",
       "      <td>-3824.021050</td>\n",
       "      <td>7.731665e+06</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.255759</td>\n",
       "      <td>7.735489e+06</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>3806389.80</td>\n",
       "      <td>1001</td>\n",
       "      <td>[{'price': 382.401105, 'order_id': 'ba457cf82d...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-20 21:00:00+00:00</th>\n",
       "      <td>395.950</td>\n",
       "      <td>0.044060</td>\n",
       "      <td>1.691163e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>36.952702</td>\n",
       "      <td>0.005180</td>\n",
       "      <td>-0.571692</td>\n",
       "      <td>-3961.489750</td>\n",
       "      <td>7.727704e+06</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.355833</td>\n",
       "      <td>7.731665e+06</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>3822100.00</td>\n",
       "      <td>1002</td>\n",
       "      <td>[{'price': 396.147975, 'order_id': '6b28a3c43a...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-21 21:00:00+00:00</th>\n",
       "      <td>396.445</td>\n",
       "      <td>0.044039</td>\n",
       "      <td>1.696116e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.091798</td>\n",
       "      <td>0.005178</td>\n",
       "      <td>-0.571615</td>\n",
       "      <td>-3966.442225</td>\n",
       "      <td>7.723737e+06</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.358742</td>\n",
       "      <td>7.727704e+06</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>3963459.50</td>\n",
       "      <td>1003</td>\n",
       "      <td>[{'price': 396.6432225, 'order_id': 'a421f95ac...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-22 21:00:00+00:00</th>\n",
       "      <td>398.550</td>\n",
       "      <td>0.044024</td>\n",
       "      <td>1.717206e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.231404</td>\n",
       "      <td>0.005175</td>\n",
       "      <td>-0.571160</td>\n",
       "      <td>-3987.502750</td>\n",
       "      <td>7.719750e+06</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.373319</td>\n",
       "      <td>7.723737e+06</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>3972378.90</td>\n",
       "      <td>1004</td>\n",
       "      <td>[{'price': 398.749275, 'order_id': 'd2fc097417...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-23 21:00:00+00:00</th>\n",
       "      <td>403.330</td>\n",
       "      <td>0.044046</td>\n",
       "      <td>1.765148e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.371522</td>\n",
       "      <td>0.005173</td>\n",
       "      <td>-0.570080</td>\n",
       "      <td>-4035.326650</td>\n",
       "      <td>7.715714e+06</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.407230</td>\n",
       "      <td>7.719750e+06</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>3997456.50</td>\n",
       "      <td>1005</td>\n",
       "      <td>[{'price': 403.531665, 'order_id': 'c93be38ceb...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-27 21:00:00+00:00</th>\n",
       "      <td>406.530</td>\n",
       "      <td>0.044043</td>\n",
       "      <td>1.797274e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.512153</td>\n",
       "      <td>0.005170</td>\n",
       "      <td>-0.569374</td>\n",
       "      <td>-4067.342650</td>\n",
       "      <td>7.711647e+06</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.429607</td>\n",
       "      <td>7.715714e+06</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>4049433.20</td>\n",
       "      <td>1006</td>\n",
       "      <td>[{'price': 406.733265, 'order_id': 'e5a6e72792...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-28 21:00:00+00:00</th>\n",
       "      <td>402.640</td>\n",
       "      <td>0.044056</td>\n",
       "      <td>1.758177e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.653300</td>\n",
       "      <td>0.005168</td>\n",
       "      <td>-0.570321</td>\n",
       "      <td>-4028.423200</td>\n",
       "      <td>7.707619e+06</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.398716</td>\n",
       "      <td>7.711647e+06</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>4085626.50</td>\n",
       "      <td>1007</td>\n",
       "      <td>[{'price': 402.84132, 'order_id': 'db8e9e4061b...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-29 21:00:00+00:00</th>\n",
       "      <td>405.120</td>\n",
       "      <td>0.044045</td>\n",
       "      <td>1.783124e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.794963</td>\n",
       "      <td>0.005165</td>\n",
       "      <td>-0.569782</td>\n",
       "      <td>-4053.235600</td>\n",
       "      <td>7.703565e+06</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.415906</td>\n",
       "      <td>7.707619e+06</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>4050558.40</td>\n",
       "      <td>1008</td>\n",
       "      <td>[{'price': 405.32256, 'order_id': '203f9f9759b...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-30 21:00:00+00:00</th>\n",
       "      <td>405.000</td>\n",
       "      <td>0.044023</td>\n",
       "      <td>1.781913e-01</td>\n",
       "      <td>0.564055</td>\n",
       "      <td>37.937147</td>\n",
       "      <td>0.005162</td>\n",
       "      <td>-0.569851</td>\n",
       "      <td>-4052.035000</td>\n",
       "      <td>7.699513e+06</td>\n",
       "      <td>4082400.00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.414337</td>\n",
       "      <td>7.703565e+06</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>4079558.40</td>\n",
       "      <td>1009</td>\n",
       "      <td>[{'price': 405.2025, 'order_id': '430b23ffa2ed...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1009 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              AAPL  algo_volatility  algorithm_period_return  \\\n",
       "2008-01-02 21:00:00+00:00  194.840              NaN             0.000000e+00   \n",
       "2008-01-03 21:00:00+00:00  194.930         0.000001            -9.846500e-08   \n",
       "2008-01-04 21:00:00+00:00  180.050         0.000137            -1.506949e-05   \n",
       "2008-01-07 21:00:00+00:00  177.640         0.000112            -1.997931e-05   \n",
       "2008-01-08 21:00:00+00:00  171.250         0.000140            -3.923593e-05   \n",
       "2008-01-09 21:00:00+00:00  179.400         0.000290            -6.726635e-06   \n",
       "2008-01-10 21:00:00+00:00  178.020         0.000267            -1.371664e-05   \n",
       "2008-01-11 21:00:00+00:00  172.690         0.000299            -4.578399e-05   \n",
       "2008-01-14 21:00:00+00:00  178.780         0.000379            -3.244380e-06   \n",
       "2008-01-15 21:00:00+00:00  169.040         0.000529            -8.124990e-05   \n",
       "2008-01-16 21:00:00+00:00  159.640         0.000621            -1.659307e-04   \n",
       "2008-01-17 21:00:00+00:00  160.890         0.000606            -1.535122e-04   \n",
       "2008-01-18 21:00:00+00:00  161.360         0.000585            -1.484238e-04   \n",
       "2008-01-22 21:00:00+00:00  155.640         0.000613            -2.171427e-04   \n",
       "2008-01-23 21:00:00+00:00  139.072         0.001010            -4.325972e-04   \n",
       "2008-01-24 21:00:00+00:00  135.600         0.000979            -4.812740e-04   \n",
       "2008-01-25 21:00:00+00:00  130.009         0.000970            -5.652050e-04   \n",
       "2008-01-28 21:00:00+00:00  130.009         0.000950            -5.652710e-04   \n",
       "2008-01-29 21:00:00+00:00  131.540         0.000946            -5.393108e-04   \n",
       "2008-01-30 21:00:00+00:00  132.180         0.000932            -5.278579e-04   \n",
       "2008-01-31 21:00:00+00:00  135.360         0.000957            -4.675065e-04   \n",
       "2008-02-01 21:00:00+00:00  133.750         0.000934            -4.997744e-04   \n",
       "2008-02-04 21:00:00+00:00  131.650         0.000915            -5.439412e-04   \n",
       "2008-02-05 21:00:00+00:00  129.360         0.000900            -5.943869e-04   \n",
       "2008-02-06 21:00:00+00:00  122.000         0.000993            -7.637289e-04   \n",
       "2008-02-07 21:00:00+00:00  121.240         0.000974            -7.820305e-04   \n",
       "2008-02-08 21:00:00+00:00  125.480         0.001042            -6.760943e-04   \n",
       "2008-02-11 21:00:00+00:00  129.449         0.001092            -5.729660e-04   \n",
       "2008-02-12 21:00:00+00:00  124.860         0.001115            -6.969324e-04   \n",
       "2008-02-13 21:00:00+00:00  129.400         0.001180            -5.698781e-04   \n",
       "...                            ...              ...                      ...   \n",
       "2011-11-17 21:00:00+00:00  377.410         0.042831             1.506754e-01   \n",
       "2011-11-18 21:00:00+00:00  374.940         0.042825             1.482571e-01   \n",
       "2011-11-21 21:00:00+00:00  369.010         0.042884             1.424455e-01   \n",
       "2011-11-22 21:00:00+00:00  376.510         0.042981             1.498029e-01   \n",
       "2011-11-23 21:00:00+00:00  366.990         0.043163             1.404540e-01   \n",
       "2011-11-25 18:00:00+00:00  363.570         0.043169             1.370920e-01   \n",
       "2011-11-28 21:00:00+00:00  376.120         0.043486             1.494410e-01   \n",
       "2011-11-29 21:00:00+00:00  373.200         0.043485             1.465646e-01   \n",
       "2011-11-30 21:00:00+00:00  382.200         0.043632             1.554384e-01   \n",
       "2011-12-01 21:00:00+00:00  387.930         0.043675             1.610937e-01   \n",
       "2011-12-02 21:00:00+00:00  389.700         0.043659             1.628423e-01   \n",
       "2011-12-05 21:00:00+00:00  393.010         0.043657             1.661157e-01   \n",
       "2011-12-06 21:00:00+00:00  390.950         0.043646             1.640761e-01   \n",
       "2011-12-07 21:00:00+00:00  389.090         0.043633             1.622326e-01   \n",
       "2011-12-08 21:00:00+00:00  390.660         0.043615             1.637899e-01   \n",
       "2011-12-09 21:00:00+00:00  393.620         0.043609             1.667290e-01   \n",
       "2011-12-12 21:00:00+00:00  391.840         0.043595             1.649594e-01   \n",
       "2011-12-13 21:00:00+00:00  388.810         0.043595             1.619444e-01   \n",
       "2011-12-14 21:00:00+00:00  380.190         0.043738             1.533587e-01   \n",
       "2011-12-15 21:00:00+00:00  378.940         0.043720             1.521122e-01   \n",
       "2011-12-16 21:00:00+00:00  381.020         0.043706             1.541879e-01   \n",
       "2011-12-19 21:00:00+00:00  382.210         0.043687             1.553765e-01   \n",
       "2011-12-20 21:00:00+00:00  395.950         0.044060             1.691163e-01   \n",
       "2011-12-21 21:00:00+00:00  396.445         0.044039             1.696116e-01   \n",
       "2011-12-22 21:00:00+00:00  398.550         0.044024             1.717206e-01   \n",
       "2011-12-23 21:00:00+00:00  403.330         0.044046             1.765148e-01   \n",
       "2011-12-27 21:00:00+00:00  406.530         0.044043             1.797274e-01   \n",
       "2011-12-28 21:00:00+00:00  402.640         0.044056             1.758177e-01   \n",
       "2011-12-29 21:00:00+00:00  405.120         0.044045             1.783124e-01   \n",
       "2011-12-30 21:00:00+00:00  405.000         0.044023             1.781913e-01   \n",
       "\n",
       "                              alpha  benchmark_period_return  \\\n",
       "2008-01-02 21:00:00+00:00       NaN                 0.003665   \n",
       "2008-01-03 21:00:00+00:00       NaN                 0.007343   \n",
       "2008-01-04 21:00:00+00:00       NaN                 0.011035   \n",
       "2008-01-07 21:00:00+00:00       NaN                 0.014741   \n",
       "2008-01-08 21:00:00+00:00       NaN                 0.018460   \n",
       "2008-01-09 21:00:00+00:00       NaN                 0.022192   \n",
       "2008-01-10 21:00:00+00:00       NaN                 0.025939   \n",
       "2008-01-11 21:00:00+00:00       NaN                 0.029699   \n",
       "2008-01-14 21:00:00+00:00       NaN                 0.033473   \n",
       "2008-01-15 21:00:00+00:00       NaN                 0.037260   \n",
       "2008-01-16 21:00:00+00:00       NaN                 0.041062   \n",
       "2008-01-17 21:00:00+00:00       NaN                 0.044877   \n",
       "2008-01-18 21:00:00+00:00       NaN                 0.048707   \n",
       "2008-01-22 21:00:00+00:00       NaN                 0.052550   \n",
       "2008-01-23 21:00:00+00:00       NaN                 0.056408   \n",
       "2008-01-24 21:00:00+00:00       NaN                 0.060280   \n",
       "2008-01-25 21:00:00+00:00       NaN                 0.064166   \n",
       "2008-01-28 21:00:00+00:00       NaN                 0.068066   \n",
       "2008-01-29 21:00:00+00:00       NaN                 0.071980   \n",
       "2008-01-30 21:00:00+00:00       NaN                 0.075909   \n",
       "2008-01-31 21:00:00+00:00       NaN                 0.079852   \n",
       "2008-02-01 21:00:00+00:00       NaN                 0.083810   \n",
       "2008-02-04 21:00:00+00:00       NaN                 0.087782   \n",
       "2008-02-05 21:00:00+00:00       NaN                 0.091769   \n",
       "2008-02-06 21:00:00+00:00       NaN                 0.095770   \n",
       "2008-02-07 21:00:00+00:00       NaN                 0.099786   \n",
       "2008-02-08 21:00:00+00:00       NaN                 0.103817   \n",
       "2008-02-11 21:00:00+00:00       NaN                 0.107862   \n",
       "2008-02-12 21:00:00+00:00       NaN                 0.111922   \n",
       "2008-02-13 21:00:00+00:00       NaN                 0.115998   \n",
       "...                             ...                      ...   \n",
       "2011-11-17 21:00:00+00:00  0.564055                34.017863   \n",
       "2011-11-18 21:00:00+00:00  0.564055                34.146203   \n",
       "2011-11-21 21:00:00+00:00  0.564055                34.275013   \n",
       "2011-11-22 21:00:00+00:00  0.564055                34.404296   \n",
       "2011-11-23 21:00:00+00:00  0.564055                34.534052   \n",
       "2011-11-25 18:00:00+00:00  0.564055                34.664284   \n",
       "2011-11-28 21:00:00+00:00  0.564055                34.794994   \n",
       "2011-11-29 21:00:00+00:00  0.564055                34.926182   \n",
       "2011-11-30 21:00:00+00:00  0.564055                35.057851   \n",
       "2011-12-01 21:00:00+00:00  0.564055                35.190003   \n",
       "2011-12-02 21:00:00+00:00  0.564055                35.322639   \n",
       "2011-12-05 21:00:00+00:00  0.564055                35.455761   \n",
       "2011-12-06 21:00:00+00:00  0.564055                35.589371   \n",
       "2011-12-07 21:00:00+00:00  0.564055                35.723471   \n",
       "2011-12-08 21:00:00+00:00  0.564055                35.858062   \n",
       "2011-12-09 21:00:00+00:00  0.564055                35.993146   \n",
       "2011-12-12 21:00:00+00:00  0.564055                36.128726   \n",
       "2011-12-13 21:00:00+00:00  0.564055                36.264802   \n",
       "2011-12-14 21:00:00+00:00  0.564055                36.401377   \n",
       "2011-12-15 21:00:00+00:00  0.564055                36.538453   \n",
       "2011-12-16 21:00:00+00:00  0.564055                36.676031   \n",
       "2011-12-19 21:00:00+00:00  0.564055                36.814113   \n",
       "2011-12-20 21:00:00+00:00  0.564055                36.952702   \n",
       "2011-12-21 21:00:00+00:00  0.564055                37.091798   \n",
       "2011-12-22 21:00:00+00:00  0.564055                37.231404   \n",
       "2011-12-23 21:00:00+00:00  0.564055                37.371522   \n",
       "2011-12-27 21:00:00+00:00  0.564055                37.512153   \n",
       "2011-12-28 21:00:00+00:00  0.564055                37.653300   \n",
       "2011-12-29 21:00:00+00:00  0.564055                37.794963   \n",
       "2011-12-30 21:00:00+00:00  0.564055                37.937147   \n",
       "\n",
       "                           benchmark_volatility      beta  capital_used  \\\n",
       "2008-01-02 21:00:00+00:00                   NaN       NaN      0.000000   \n",
       "2008-01-03 21:00:00+00:00              0.000000       NaN  -1950.284650   \n",
       "2008-01-04 21:00:00+00:00              0.000000       NaN  -1801.410250   \n",
       "2008-01-07 21:00:00+00:00              0.000000       NaN  -1777.298200   \n",
       "2008-01-08 21:00:00+00:00              0.000000       NaN  -1713.366250   \n",
       "2008-01-09 21:00:00+00:00              0.000000       NaN  -1794.907000   \n",
       "2008-01-10 21:00:00+00:00              0.000000       NaN  -1781.100100   \n",
       "2008-01-11 21:00:00+00:00              0.000000       NaN  -1727.773450   \n",
       "2008-01-14 21:00:00+00:00              0.000000       NaN  -1788.703900   \n",
       "2008-01-15 21:00:00+00:00              0.000000       NaN  -1691.255200   \n",
       "2008-01-16 21:00:00+00:00              0.000000       NaN  -1597.208200   \n",
       "2008-01-17 21:00:00+00:00              0.000000       NaN  -1609.714450   \n",
       "2008-01-18 21:00:00+00:00              0.000000       NaN  -1614.416800   \n",
       "2008-01-22 21:00:00+00:00              0.000000       NaN  -1557.188200   \n",
       "2008-01-23 21:00:00+00:00              0.000000       NaN  -1391.425360   \n",
       "2008-01-24 21:00:00+00:00              0.000000       NaN  -1356.688000   \n",
       "2008-01-25 21:00:00+00:00              0.000000       NaN  -1300.750045   \n",
       "2008-01-28 21:00:00+00:00              0.000000       NaN  -1300.750045   \n",
       "2008-01-29 21:00:00+00:00              0.000000       NaN  -1316.067700   \n",
       "2008-01-30 21:00:00+00:00              0.000000       NaN  -1322.470900   \n",
       "2008-01-31 21:00:00+00:00              0.000000       NaN  -1354.286800   \n",
       "2008-02-01 21:00:00+00:00              0.000000       NaN  -1338.178750   \n",
       "2008-02-04 21:00:00+00:00              0.000000       NaN  -1317.168250   \n",
       "2008-02-05 21:00:00+00:00              0.000000       NaN  -1294.256800   \n",
       "2008-02-06 21:00:00+00:00              0.000000       NaN  -1220.620000   \n",
       "2008-02-07 21:00:00+00:00              0.000000       NaN  -1213.016200   \n",
       "2008-02-08 21:00:00+00:00              0.000000       NaN  -1255.437400   \n",
       "2008-02-11 21:00:00+00:00              0.000000       NaN  -1295.147245   \n",
       "2008-02-12 21:00:00+00:00              0.000000       NaN  -1249.234300   \n",
       "2008-02-13 21:00:00+00:00              0.000000       NaN  -1294.657000   \n",
       "...                                         ...       ...           ...   \n",
       "2011-11-17 21:00:00+00:00              0.005238 -0.575354  -3775.997050   \n",
       "2011-11-18 21:00:00+00:00              0.005235 -0.575979  -3751.284700   \n",
       "2011-11-21 21:00:00+00:00              0.005232 -0.577433  -3691.955050   \n",
       "2011-11-22 21:00:00+00:00              0.005230 -0.575665  -3766.992550   \n",
       "2011-11-23 21:00:00+00:00              0.005227 -0.577974  -3671.744950   \n",
       "2011-11-25 18:00:00+00:00              0.005225 -0.578831  -3637.527850   \n",
       "2011-11-28 21:00:00+00:00              0.005222 -0.575833  -3763.090600   \n",
       "2011-11-29 21:00:00+00:00              0.005219 -0.576566  -3733.876000   \n",
       "2011-11-30 21:00:00+00:00              0.005217 -0.574446  -3823.921000   \n",
       "2011-12-01 21:00:00+00:00              0.005214 -0.573122  -3881.249650   \n",
       "2011-12-02 21:00:00+00:00              0.005211 -0.572742  -3898.958500   \n",
       "2011-12-05 21:00:00+00:00              0.005209 -0.571999  -3932.075050   \n",
       "2011-12-06 21:00:00+00:00              0.005206 -0.572524  -3911.464750   \n",
       "2011-12-07 21:00:00+00:00              0.005204 -0.573001  -3892.855450   \n",
       "2011-12-08 21:00:00+00:00              0.005201 -0.572668  -3908.563300   \n",
       "2011-12-09 21:00:00+00:00              0.005198 -0.572009  -3938.178100   \n",
       "2011-12-12 21:00:00+00:00              0.005196 -0.572467  -3920.369200   \n",
       "2011-12-13 21:00:00+00:00              0.005193 -0.573220  -3890.054050   \n",
       "2011-12-14 21:00:00+00:00              0.005191 -0.575294  -3803.810950   \n",
       "2011-12-15 21:00:00+00:00              0.005188 -0.575627  -3791.304700   \n",
       "2011-12-16 21:00:00+00:00              0.005186 -0.575167  -3812.115100   \n",
       "2011-12-19 21:00:00+00:00              0.005183 -0.574920  -3824.021050   \n",
       "2011-12-20 21:00:00+00:00              0.005180 -0.571692  -3961.489750   \n",
       "2011-12-21 21:00:00+00:00              0.005178 -0.571615  -3966.442225   \n",
       "2011-12-22 21:00:00+00:00              0.005175 -0.571160  -3987.502750   \n",
       "2011-12-23 21:00:00+00:00              0.005173 -0.570080  -4035.326650   \n",
       "2011-12-27 21:00:00+00:00              0.005170 -0.569374  -4067.342650   \n",
       "2011-12-28 21:00:00+00:00              0.005168 -0.570321  -4028.423200   \n",
       "2011-12-29 21:00:00+00:00              0.005165 -0.569782  -4053.235600   \n",
       "2011-12-30 21:00:00+00:00              0.005162 -0.569851  -4052.035000   \n",
       "\n",
       "                            ending_cash  ending_exposure  \\\n",
       "2008-01-02 21:00:00+00:00  1.000000e+07             0.00   \n",
       "2008-01-03 21:00:00+00:00  9.998050e+06          1949.30   \n",
       "2008-01-04 21:00:00+00:00  9.996248e+06          3601.00   \n",
       "2008-01-07 21:00:00+00:00  9.994471e+06          5329.20   \n",
       "2008-01-08 21:00:00+00:00  9.992758e+06          6850.00   \n",
       "2008-01-09 21:00:00+00:00  9.990963e+06          8970.00   \n",
       "2008-01-10 21:00:00+00:00  9.989182e+06         10681.20   \n",
       "2008-01-11 21:00:00+00:00  9.987454e+06         12088.30   \n",
       "2008-01-14 21:00:00+00:00  9.985665e+06         14302.40   \n",
       "2008-01-15 21:00:00+00:00  9.983974e+06         15213.60   \n",
       "2008-01-16 21:00:00+00:00  9.982377e+06         15964.00   \n",
       "2008-01-17 21:00:00+00:00  9.980767e+06         17697.90   \n",
       "2008-01-18 21:00:00+00:00  9.979153e+06         19363.20   \n",
       "2008-01-22 21:00:00+00:00  9.977595e+06         20233.20   \n",
       "2008-01-23 21:00:00+00:00  9.976204e+06         19470.08   \n",
       "2008-01-24 21:00:00+00:00  9.974847e+06         20340.00   \n",
       "2008-01-25 21:00:00+00:00  9.973547e+06         20801.44   \n",
       "2008-01-28 21:00:00+00:00  9.972246e+06         22101.53   \n",
       "2008-01-29 21:00:00+00:00  9.970930e+06         23677.20   \n",
       "2008-01-30 21:00:00+00:00  9.969607e+06         25114.20   \n",
       "2008-01-31 21:00:00+00:00  9.968253e+06         27072.00   \n",
       "2008-02-01 21:00:00+00:00  9.966915e+06         28087.50   \n",
       "2008-02-04 21:00:00+00:00  9.965598e+06         28963.00   \n",
       "2008-02-05 21:00:00+00:00  9.964303e+06         29752.80   \n",
       "2008-02-06 21:00:00+00:00  9.963083e+06         29280.00   \n",
       "2008-02-07 21:00:00+00:00  9.961870e+06         30310.00   \n",
       "2008-02-08 21:00:00+00:00  9.960614e+06         32624.80   \n",
       "2008-02-11 21:00:00+00:00  9.959319e+06         34951.23   \n",
       "2008-02-12 21:00:00+00:00  9.958070e+06         34960.80   \n",
       "2008-02-13 21:00:00+00:00  9.956775e+06         37526.00   \n",
       "...                                 ...              ...   \n",
       "2011-11-17 21:00:00+00:00  7.811911e+06       3694843.90   \n",
       "2011-11-18 21:00:00+00:00  7.808159e+06       3674412.00   \n",
       "2011-11-21 21:00:00+00:00  7.804467e+06       3619988.10   \n",
       "2011-11-22 21:00:00+00:00  7.800700e+06       3697328.20   \n",
       "2011-11-23 21:00:00+00:00  7.797029e+06       3607511.70   \n",
       "2011-11-25 18:00:00+00:00  7.793391e+06       3577528.80   \n",
       "2011-11-28 21:00:00+00:00  7.789628e+06       3704782.00   \n",
       "2011-11-29 21:00:00+00:00  7.785894e+06       3679752.00   \n",
       "2011-11-30 21:00:00+00:00  7.782070e+06       3772314.00   \n",
       "2011-12-01 21:00:00+00:00  7.778189e+06       3832748.40   \n",
       "2011-12-02 21:00:00+00:00  7.774290e+06       3854133.00   \n",
       "2011-12-05 21:00:00+00:00  7.770358e+06       3890799.00   \n",
       "2011-12-06 21:00:00+00:00  7.766446e+06       3874314.50   \n",
       "2011-12-07 21:00:00+00:00  7.762554e+06       3859772.80   \n",
       "2011-12-08 21:00:00+00:00  7.758645e+06       3879253.80   \n",
       "2011-12-09 21:00:00+00:00  7.754707e+06       3912582.80   \n",
       "2011-12-12 21:00:00+00:00  7.750786e+06       3898808.00   \n",
       "2011-12-13 21:00:00+00:00  7.746896e+06       3872547.60   \n",
       "2011-12-14 21:00:00+00:00  7.743093e+06       3790494.30   \n",
       "2011-12-15 21:00:00+00:00  7.739301e+06       3781821.20   \n",
       "2011-12-16 21:00:00+00:00  7.735489e+06       3806389.80   \n",
       "2011-12-19 21:00:00+00:00  7.731665e+06       3822100.00   \n",
       "2011-12-20 21:00:00+00:00  7.727704e+06       3963459.50   \n",
       "2011-12-21 21:00:00+00:00  7.723737e+06       3972378.90   \n",
       "2011-12-22 21:00:00+00:00  7.719750e+06       3997456.50   \n",
       "2011-12-23 21:00:00+00:00  7.715714e+06       4049433.20   \n",
       "2011-12-27 21:00:00+00:00  7.711647e+06       4085626.50   \n",
       "2011-12-28 21:00:00+00:00  7.707619e+06       4050558.40   \n",
       "2011-12-29 21:00:00+00:00  7.703565e+06       4079558.40   \n",
       "2011-12-30 21:00:00+00:00  7.699513e+06       4082400.00   \n",
       "\n",
       "                                   ...            short_exposure  short_value  \\\n",
       "2008-01-02 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-03 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-04 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-09 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-10 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-11 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-14 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-15 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-16 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-17 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-18 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-24 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-25 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-01-31 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-01 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-04 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-05 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-06 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-11 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-12 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2008-02-13 21:00:00+00:00          ...                       0.0          0.0   \n",
       "...                                ...                       ...          ...   \n",
       "2011-11-17 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-18 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-21 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-25 18:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-11-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-01 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-02 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-05 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-06 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-07 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-08 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-09 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-12 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-13 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-14 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-15 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-16 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-19 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-20 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-21 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-22 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-23 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-27 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-28 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-29 21:00:00+00:00          ...                       0.0          0.0   \n",
       "2011-12-30 21:00:00+00:00          ...                       0.0          0.0   \n",
       "\n",
       "                           shorts_count    sortino  starting_cash  \\\n",
       "2008-01-02 21:00:00+00:00             0        NaN   1.000000e+07   \n",
       "2008-01-03 21:00:00+00:00             0 -11.224972   1.000000e+07   \n",
       "2008-01-04 21:00:00+00:00             0  -9.225231   9.998050e+06   \n",
       "2008-01-07 21:00:00+00:00             0 -10.064893   9.996248e+06   \n",
       "2008-01-08 21:00:00+00:00             0 -11.195170   9.994471e+06   \n",
       "2008-01-09 21:00:00+00:00             0  -1.751847   9.992758e+06   \n",
       "2008-01-10 21:00:00+00:00             0  -3.184238   9.990963e+06   \n",
       "2008-01-11 21:00:00+00:00             0  -6.239021   9.989182e+06   \n",
       "2008-01-14 21:00:00+00:00             0  -0.416540   9.987454e+06   \n",
       "2008-01-15 21:00:00+00:00             0  -4.623698   9.985665e+06   \n",
       "2008-01-16 21:00:00+00:00             0  -6.494960   9.983974e+06   \n",
       "2008-01-17 21:00:00+00:00             0  -5.752987   9.982377e+06   \n",
       "2008-01-18 21:00:00+00:00             0  -5.344059   9.980767e+06   \n",
       "2008-01-22 21:00:00+00:00             0  -6.567920   9.979153e+06   \n",
       "2008-01-23 21:00:00+00:00             0  -6.896659   9.977595e+06   \n",
       "2008-01-24 21:00:00+00:00             0  -7.299514   9.976204e+06   \n",
       "2008-01-25 21:00:00+00:00             0  -7.919264   9.974847e+06   \n",
       "2008-01-28 21:00:00+00:00             0  -7.697040   9.973547e+06   \n",
       "2008-01-29 21:00:00+00:00             0  -7.147567   9.972246e+06   \n",
       "2008-01-30 21:00:00+00:00             0  -6.818592   9.970930e+06   \n",
       "2008-01-31 21:00:00+00:00             0  -5.893207   9.969607e+06   \n",
       "2008-02-01 21:00:00+00:00             0  -6.113220   9.968253e+06   \n",
       "2008-02-04 21:00:00+00:00             0  -6.425988   9.966915e+06   \n",
       "2008-02-05 21:00:00+00:00             0  -6.765445   9.965598e+06   \n",
       "2008-02-06 21:00:00+00:00             0  -7.320042   9.964303e+06   \n",
       "2008-02-07 21:00:00+00:00             0  -7.338775   9.963083e+06   \n",
       "2008-02-08 21:00:00+00:00             0  -6.225585   9.961870e+06   \n",
       "2008-02-11 21:00:00+00:00             0  -5.180489   9.960614e+06   \n",
       "2008-02-12 21:00:00+00:00             0  -5.800272   9.959319e+06   \n",
       "2008-02-13 21:00:00+00:00             0  -4.662658   9.958070e+06   \n",
       "...                                 ...        ...            ...   \n",
       "2011-11-17 21:00:00+00:00             0   1.265484   7.815687e+06   \n",
       "2011-11-18 21:00:00+00:00             0   1.245542   7.811911e+06   \n",
       "2011-11-21 21:00:00+00:00             0   1.195890   7.808159e+06   \n",
       "2011-11-22 21:00:00+00:00             0   1.251567   7.804467e+06   \n",
       "2011-11-23 21:00:00+00:00             0   1.168451   7.800700e+06   \n",
       "2011-11-25 18:00:00+00:00             0   1.140921   7.797029e+06   \n",
       "2011-11-28 21:00:00+00:00             0   1.234077   7.793391e+06   \n",
       "2011-11-29 21:00:00+00:00             0   1.210759   7.789628e+06   \n",
       "2011-11-30 21:00:00+00:00             0   1.276818   7.785894e+06   \n",
       "2011-12-01 21:00:00+00:00             0   1.318314   7.782070e+06   \n",
       "2011-12-02 21:00:00+00:00             0   1.330608   7.778189e+06   \n",
       "2011-12-05 21:00:00+00:00             0   1.354150   7.774290e+06   \n",
       "2011-12-06 21:00:00+00:00             0   1.337835   7.770358e+06   \n",
       "2011-12-07 21:00:00+00:00             0   1.323077   7.766446e+06   \n",
       "2011-12-08 21:00:00+00:00             0   1.333910   7.762554e+06   \n",
       "2011-12-09 21:00:00+00:00             0   1.354900   7.758645e+06   \n",
       "2011-12-12 21:00:00+00:00             0   1.340770   7.754707e+06   \n",
       "2011-12-13 21:00:00+00:00             0   1.316649   7.750786e+06   \n",
       "2011-12-14 21:00:00+00:00             0   1.242998   7.746896e+06   \n",
       "2011-12-15 21:00:00+00:00             0   1.233004   7.743093e+06   \n",
       "2011-12-16 21:00:00+00:00             0   1.247658   7.739301e+06   \n",
       "2011-12-19 21:00:00+00:00             0   1.255759   7.735489e+06   \n",
       "2011-12-20 21:00:00+00:00             0   1.355833   7.731665e+06   \n",
       "2011-12-21 21:00:00+00:00             0   1.358742   7.727704e+06   \n",
       "2011-12-22 21:00:00+00:00             0   1.373319   7.723737e+06   \n",
       "2011-12-23 21:00:00+00:00             0   1.407230   7.719750e+06   \n",
       "2011-12-27 21:00:00+00:00             0   1.429607   7.715714e+06   \n",
       "2011-12-28 21:00:00+00:00             0   1.398716   7.711647e+06   \n",
       "2011-12-29 21:00:00+00:00             0   1.415906   7.707619e+06   \n",
       "2011-12-30 21:00:00+00:00             0   1.414337   7.703565e+06   \n",
       "\n",
       "                           starting_exposure  starting_value  trading_days  \\\n",
       "2008-01-02 21:00:00+00:00               0.00            0.00             1   \n",
       "2008-01-03 21:00:00+00:00               0.00            0.00             2   \n",
       "2008-01-04 21:00:00+00:00            1949.30         1949.30             3   \n",
       "2008-01-07 21:00:00+00:00            3601.00         3601.00             4   \n",
       "2008-01-08 21:00:00+00:00            5329.20         5329.20             5   \n",
       "2008-01-09 21:00:00+00:00            6850.00         6850.00             6   \n",
       "2008-01-10 21:00:00+00:00            8970.00         8970.00             7   \n",
       "2008-01-11 21:00:00+00:00           10681.20        10681.20             8   \n",
       "2008-01-14 21:00:00+00:00           12088.30        12088.30             9   \n",
       "2008-01-15 21:00:00+00:00           14302.40        14302.40            10   \n",
       "2008-01-16 21:00:00+00:00           15213.60        15213.60            11   \n",
       "2008-01-17 21:00:00+00:00           15964.00        15964.00            12   \n",
       "2008-01-18 21:00:00+00:00           17697.90        17697.90            13   \n",
       "2008-01-22 21:00:00+00:00           19363.20        19363.20            14   \n",
       "2008-01-23 21:00:00+00:00           20233.20        20233.20            15   \n",
       "2008-01-24 21:00:00+00:00           19470.08        19470.08            16   \n",
       "2008-01-25 21:00:00+00:00           20340.00        20340.00            17   \n",
       "2008-01-28 21:00:00+00:00           20801.44        20801.44            18   \n",
       "2008-01-29 21:00:00+00:00           22101.53        22101.53            19   \n",
       "2008-01-30 21:00:00+00:00           23677.20        23677.20            20   \n",
       "2008-01-31 21:00:00+00:00           25114.20        25114.20            21   \n",
       "2008-02-01 21:00:00+00:00           27072.00        27072.00            22   \n",
       "2008-02-04 21:00:00+00:00           28087.50        28087.50            23   \n",
       "2008-02-05 21:00:00+00:00           28963.00        28963.00            24   \n",
       "2008-02-06 21:00:00+00:00           29752.80        29752.80            25   \n",
       "2008-02-07 21:00:00+00:00           29280.00        29280.00            26   \n",
       "2008-02-08 21:00:00+00:00           30310.00        30310.00            27   \n",
       "2008-02-11 21:00:00+00:00           32624.80        32624.80            28   \n",
       "2008-02-12 21:00:00+00:00           34951.23        34951.23            29   \n",
       "2008-02-13 21:00:00+00:00           34960.80        34960.80            30   \n",
       "...                                      ...             ...           ...   \n",
       "2011-11-17 21:00:00+00:00         3763050.60      3763050.60           980   \n",
       "2011-11-18 21:00:00+00:00         3694843.90      3694843.90           981   \n",
       "2011-11-21 21:00:00+00:00         3674412.00      3674412.00           982   \n",
       "2011-11-22 21:00:00+00:00         3619988.10      3619988.10           983   \n",
       "2011-11-23 21:00:00+00:00         3697328.20      3697328.20           984   \n",
       "2011-11-25 18:00:00+00:00         3607511.70      3607511.70           985   \n",
       "2011-11-28 21:00:00+00:00         3577528.80      3577528.80           986   \n",
       "2011-11-29 21:00:00+00:00         3704782.00      3704782.00           987   \n",
       "2011-11-30 21:00:00+00:00         3679752.00      3679752.00           988   \n",
       "2011-12-01 21:00:00+00:00         3772314.00      3772314.00           989   \n",
       "2011-12-02 21:00:00+00:00         3832748.40      3832748.40           990   \n",
       "2011-12-05 21:00:00+00:00         3854133.00      3854133.00           991   \n",
       "2011-12-06 21:00:00+00:00         3890799.00      3890799.00           992   \n",
       "2011-12-07 21:00:00+00:00         3874314.50      3874314.50           993   \n",
       "2011-12-08 21:00:00+00:00         3859772.80      3859772.80           994   \n",
       "2011-12-09 21:00:00+00:00         3879253.80      3879253.80           995   \n",
       "2011-12-12 21:00:00+00:00         3912582.80      3912582.80           996   \n",
       "2011-12-13 21:00:00+00:00         3898808.00      3898808.00           997   \n",
       "2011-12-14 21:00:00+00:00         3872547.60      3872547.60           998   \n",
       "2011-12-15 21:00:00+00:00         3790494.30      3790494.30           999   \n",
       "2011-12-16 21:00:00+00:00         3781821.20      3781821.20          1000   \n",
       "2011-12-19 21:00:00+00:00         3806389.80      3806389.80          1001   \n",
       "2011-12-20 21:00:00+00:00         3822100.00      3822100.00          1002   \n",
       "2011-12-21 21:00:00+00:00         3963459.50      3963459.50          1003   \n",
       "2011-12-22 21:00:00+00:00         3972378.90      3972378.90          1004   \n",
       "2011-12-23 21:00:00+00:00         3997456.50      3997456.50          1005   \n",
       "2011-12-27 21:00:00+00:00         4049433.20      4049433.20          1006   \n",
       "2011-12-28 21:00:00+00:00         4085626.50      4085626.50          1007   \n",
       "2011-12-29 21:00:00+00:00         4050558.40      4050558.40          1008   \n",
       "2011-12-30 21:00:00+00:00         4079558.40      4079558.40          1009   \n",
       "\n",
       "                                                                transactions  \\\n",
       "2008-01-02 21:00:00+00:00                                                 []   \n",
       "2008-01-03 21:00:00+00:00  [{'price': 195.027465, 'order_id': 'ec52ad7022...   \n",
       "2008-01-04 21:00:00+00:00  [{'price': 180.140025, 'order_id': '6448608787...   \n",
       "2008-01-07 21:00:00+00:00  [{'price': 177.72882, 'order_id': '24bc1361705...   \n",
       "2008-01-08 21:00:00+00:00  [{'price': 171.335625, 'order_id': 'cd411aad25...   \n",
       "2008-01-09 21:00:00+00:00  [{'price': 179.4897, 'order_id': '2b3a53bd317c...   \n",
       "2008-01-10 21:00:00+00:00  [{'price': 178.10901, 'order_id': '4e25a9996e6...   \n",
       "2008-01-11 21:00:00+00:00  [{'price': 172.776345, 'order_id': '39f18b2e27...   \n",
       "2008-01-14 21:00:00+00:00  [{'price': 178.86939, 'order_id': 'dc8a32a5821...   \n",
       "2008-01-15 21:00:00+00:00  [{'price': 169.12452, 'order_id': '2fc850446ed...   \n",
       "2008-01-16 21:00:00+00:00  [{'price': 159.71982000000003, 'order_id': '7e...   \n",
       "2008-01-17 21:00:00+00:00  [{'price': 160.970445, 'order_id': 'bb06e2249a...   \n",
       "2008-01-18 21:00:00+00:00  [{'price': 161.44068000000001, 'order_id': 'df...   \n",
       "2008-01-22 21:00:00+00:00  [{'price': 155.71782000000002, 'order_id': '3a...   \n",
       "2008-01-23 21:00:00+00:00  [{'price': 139.141536, 'order_id': '4dc593488f...   \n",
       "2008-01-24 21:00:00+00:00  [{'price': 135.6678, 'order_id': '95328d0e66bc...   \n",
       "2008-01-25 21:00:00+00:00  [{'price': 130.0740045, 'order_id': 'aff646a2c...   \n",
       "2008-01-28 21:00:00+00:00  [{'price': 130.0740045, 'order_id': '983ef46c1...   \n",
       "2008-01-29 21:00:00+00:00  [{'price': 131.60576999999998, 'order_id': 'a6...   \n",
       "2008-01-30 21:00:00+00:00  [{'price': 132.24609, 'order_id': '3f20771de51...   \n",
       "2008-01-31 21:00:00+00:00  [{'price': 135.42768, 'order_id': '372d6c56fd5...   \n",
       "2008-02-01 21:00:00+00:00  [{'price': 133.816875, 'order_id': '5a9d61f73b...   \n",
       "2008-02-04 21:00:00+00:00  [{'price': 131.715825, 'order_id': '3ff3b04eaf...   \n",
       "2008-02-05 21:00:00+00:00  [{'price': 129.42468000000002, 'order_id': '25...   \n",
       "2008-02-06 21:00:00+00:00  [{'price': 122.061, 'order_id': '1bc784b04e2b4...   \n",
       "2008-02-07 21:00:00+00:00  [{'price': 121.30062000000001, 'order_id': 'c6...   \n",
       "2008-02-08 21:00:00+00:00  [{'price': 125.54274000000001, 'order_id': 'b7...   \n",
       "2008-02-11 21:00:00+00:00  [{'price': 129.51372450000002, 'order_id': '92...   \n",
       "2008-02-12 21:00:00+00:00  [{'price': 124.92243, 'order_id': '194f18a4306...   \n",
       "2008-02-13 21:00:00+00:00  [{'price': 129.4647, 'order_id': '7a1b6925d46d...   \n",
       "...                                                                      ...   \n",
       "2011-11-17 21:00:00+00:00  [{'price': 377.59870500000005, 'order_id': '8c...   \n",
       "2011-11-18 21:00:00+00:00  [{'price': 375.12747, 'order_id': '44182aeb52b...   \n",
       "2011-11-21 21:00:00+00:00  [{'price': 369.194505, 'order_id': 'a41897b256...   \n",
       "2011-11-22 21:00:00+00:00  [{'price': 376.698255, 'order_id': 'c4ad06aee8...   \n",
       "2011-11-23 21:00:00+00:00  [{'price': 367.173495, 'order_id': '108fae2a48...   \n",
       "2011-11-25 18:00:00+00:00  [{'price': 363.751785, 'order_id': '50af5af4c1...   \n",
       "2011-11-28 21:00:00+00:00  [{'price': 376.30806, 'order_id': '7e9422a0925...   \n",
       "2011-11-29 21:00:00+00:00  [{'price': 373.3866, 'order_id': '7be7f3afe333...   \n",
       "2011-11-30 21:00:00+00:00  [{'price': 382.3911, 'order_id': '8257dea0cf46...   \n",
       "2011-12-01 21:00:00+00:00  [{'price': 388.123965, 'order_id': '410334c727...   \n",
       "2011-12-02 21:00:00+00:00  [{'price': 389.89484999999996, 'order_id': 'ad...   \n",
       "2011-12-05 21:00:00+00:00  [{'price': 393.206505, 'order_id': 'cdb699ea05...   \n",
       "2011-12-06 21:00:00+00:00  [{'price': 391.145475, 'order_id': '8686544864...   \n",
       "2011-12-07 21:00:00+00:00  [{'price': 389.28454500000004, 'order_id': '3b...   \n",
       "2011-12-08 21:00:00+00:00  [{'price': 390.85533000000004, 'order_id': '78...   \n",
       "2011-12-09 21:00:00+00:00  [{'price': 393.81681000000003, 'order_id': 'c0...   \n",
       "2011-12-12 21:00:00+00:00  [{'price': 392.03592000000003, 'order_id': '39...   \n",
       "2011-12-13 21:00:00+00:00  [{'price': 389.004405, 'order_id': '5fa77db006...   \n",
       "2011-12-14 21:00:00+00:00  [{'price': 380.380095, 'order_id': '8ce63b39bf...   \n",
       "2011-12-15 21:00:00+00:00  [{'price': 379.12946999999997, 'order_id': '67...   \n",
       "2011-12-16 21:00:00+00:00  [{'price': 381.21051, 'order_id': 'a814483b0ed...   \n",
       "2011-12-19 21:00:00+00:00  [{'price': 382.401105, 'order_id': 'ba457cf82d...   \n",
       "2011-12-20 21:00:00+00:00  [{'price': 396.147975, 'order_id': '6b28a3c43a...   \n",
       "2011-12-21 21:00:00+00:00  [{'price': 396.6432225, 'order_id': 'a421f95ac...   \n",
       "2011-12-22 21:00:00+00:00  [{'price': 398.749275, 'order_id': 'd2fc097417...   \n",
       "2011-12-23 21:00:00+00:00  [{'price': 403.531665, 'order_id': 'c93be38ceb...   \n",
       "2011-12-27 21:00:00+00:00  [{'price': 406.733265, 'order_id': 'e5a6e72792...   \n",
       "2011-12-28 21:00:00+00:00  [{'price': 402.84132, 'order_id': 'db8e9e4061b...   \n",
       "2011-12-29 21:00:00+00:00  [{'price': 405.32256, 'order_id': '203f9f9759b...   \n",
       "2011-12-30 21:00:00+00:00  [{'price': 405.2025, 'order_id': '430b23ffa2ed...   \n",
       "\n",
       "                          treasury_period_return  \n",
       "2008-01-02 21:00:00+00:00                    0.0  \n",
       "2008-01-03 21:00:00+00:00                    0.0  \n",
       "2008-01-04 21:00:00+00:00                    0.0  \n",
       "2008-01-07 21:00:00+00:00                    0.0  \n",
       "2008-01-08 21:00:00+00:00                    0.0  \n",
       "2008-01-09 21:00:00+00:00                    0.0  \n",
       "2008-01-10 21:00:00+00:00                    0.0  \n",
       "2008-01-11 21:00:00+00:00                    0.0  \n",
       "2008-01-14 21:00:00+00:00                    0.0  \n",
       "2008-01-15 21:00:00+00:00                    0.0  \n",
       "2008-01-16 21:00:00+00:00                    0.0  \n",
       "2008-01-17 21:00:00+00:00                    0.0  \n",
       "2008-01-18 21:00:00+00:00                    0.0  \n",
       "2008-01-22 21:00:00+00:00                    0.0  \n",
       "2008-01-23 21:00:00+00:00                    0.0  \n",
       "2008-01-24 21:00:00+00:00                    0.0  \n",
       "2008-01-25 21:00:00+00:00                    0.0  \n",
       "2008-01-28 21:00:00+00:00                    0.0  \n",
       "2008-01-29 21:00:00+00:00                    0.0  \n",
       "2008-01-30 21:00:00+00:00                    0.0  \n",
       "2008-01-31 21:00:00+00:00                    0.0  \n",
       "2008-02-01 21:00:00+00:00                    0.0  \n",
       "2008-02-04 21:00:00+00:00                    0.0  \n",
       "2008-02-05 21:00:00+00:00                    0.0  \n",
       "2008-02-06 21:00:00+00:00                    0.0  \n",
       "2008-02-07 21:00:00+00:00                    0.0  \n",
       "2008-02-08 21:00:00+00:00                    0.0  \n",
       "2008-02-11 21:00:00+00:00                    0.0  \n",
       "2008-02-12 21:00:00+00:00                    0.0  \n",
       "2008-02-13 21:00:00+00:00                    0.0  \n",
       "...                                          ...  \n",
       "2011-11-17 21:00:00+00:00                    0.0  \n",
       "2011-11-18 21:00:00+00:00                    0.0  \n",
       "2011-11-21 21:00:00+00:00                    0.0  \n",
       "2011-11-22 21:00:00+00:00                    0.0  \n",
       "2011-11-23 21:00:00+00:00                    0.0  \n",
       "2011-11-25 18:00:00+00:00                    0.0  \n",
       "2011-11-28 21:00:00+00:00                    0.0  \n",
       "2011-11-29 21:00:00+00:00                    0.0  \n",
       "2011-11-30 21:00:00+00:00                    0.0  \n",
       "2011-12-01 21:00:00+00:00                    0.0  \n",
       "2011-12-02 21:00:00+00:00                    0.0  \n",
       "2011-12-05 21:00:00+00:00                    0.0  \n",
       "2011-12-06 21:00:00+00:00                    0.0  \n",
       "2011-12-07 21:00:00+00:00                    0.0  \n",
       "2011-12-08 21:00:00+00:00                    0.0  \n",
       "2011-12-09 21:00:00+00:00                    0.0  \n",
       "2011-12-12 21:00:00+00:00                    0.0  \n",
       "2011-12-13 21:00:00+00:00                    0.0  \n",
       "2011-12-14 21:00:00+00:00                    0.0  \n",
       "2011-12-15 21:00:00+00:00                    0.0  \n",
       "2011-12-16 21:00:00+00:00                    0.0  \n",
       "2011-12-19 21:00:00+00:00                    0.0  \n",
       "2011-12-20 21:00:00+00:00                    0.0  \n",
       "2011-12-21 21:00:00+00:00                    0.0  \n",
       "2011-12-22 21:00:00+00:00                    0.0  \n",
       "2011-12-23 21:00:00+00:00                    0.0  \n",
       "2011-12-27 21:00:00+00:00                    0.0  \n",
       "2011-12-28 21:00:00+00:00                    0.0  \n",
       "2011-12-29 21:00:00+00:00                    0.0  \n",
       "2011-12-30 21:00:00+00:00                    0.0  \n",
       "\n",
       "[1009 rows x 38 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Collect Apple's share price data from 2008 to 2012, using Zipline\n",
    "%zipline --bundle quantopian-quandl --start 2008-1-1 --end 2012-1-1 -o strat.pickle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"customdat\"></a> Custom Data Panel\n",
    "\n",
    "In this section we will be working with our own __custom data__. With Zipline, we will learn here how to run backtests on our own custom csv datafiles.\n",
    "\n",
    "A large caveat here though, is that you can only use less data than how much RAM you have available. In other words, if you only have 3 GB of RAM, yet your datafile is 100 GB large - you will be unable to handle it with this method.\n",
    "\n",
    "Multiple pandas dataframes can be combined into a _depth_, using __[pandas panels](https://pandas.pydata.org/pandas-docs/stable/generated/pandas.Panel.html)__. These can be thought of as __3D dataframe__, where you have your single dataframes (which are columns and rows) and the _depth_ part of the panel is the different ticker symbols (in this case) that those rows represent.  \n",
    "\n",
    "Here we will first load in our csv files to be put into a pandas panel. The csv files must be arranged to have the columns from left to right: ```date,open,high,low,close,adj close, volume```. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from collections import OrderedDict\n",
    "import pytz  # Time-zones "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               open     high    close   volume\n",
      "date                                          \n",
      "1993-01-29  43.9687  43.9687  43.9375  1003200\n",
      "1993-02-01  43.9687  44.2500  44.2500   480500\n",
      "1993-02-02  44.2187  44.3750  44.3437   201300\n",
      "1993-02-03  44.4062  44.8437  44.8125   529400\n",
      "1993-02-04  44.9687  45.0937  45.0000   531500\n"
     ]
    }
   ],
   "source": [
    "# Load data: S&P 500 ETF daily prices\n",
    "data = OrderedDict()  # Will contain all of our data frames\n",
    "tickers = ['SPY']  # S&P 500 ETF ticker\n",
    "\n",
    "for ticker in tickers:  # Load data for all specified tickers \n",
    "    data[ticker] = pd.read_csv('{}.csv'.format(ticker), index_col=0, parse_dates=['date'])\n",
    "    data[ticker] = data[ticker][['open','high','close','volume']]  # Remove adj close column\n",
    "    print(data[ticker].head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.panel.Panel'>\n",
      "Dimensions: 1 (items) x 6329 (major_axis) x 4 (minor_axis)\n",
      "Items axis: SPY to SPY\n",
      "Major_axis axis: 1993-01-29 00:00:00+00:00 to 2018-03-16 00:00:00+00:00\n",
      "Minor_axis axis: open to volume\n"
     ]
    }
   ],
   "source": [
    "# Convert our dataframe into a panel, and adjust time zone\n",
    "panel = pd.Panel(data)\n",
    "panel.minor_axis = ['open','high','close','volume']\n",
    "panel.major_axis = panel.major_axis.tz_localize(pytz.utc)  # Convert time zone: UTC\n",
    "print(panel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies\n",
    "from zipline.api import order, record, symbol, set_benchmark\n",
    "import zipline\n",
    "from datetime import datetime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Typical Quantopian default functions, used with Zipline\n",
    "def initialize(context):\n",
    "    set_benchmark(symbol('SPY'))  # Our custom data: S&P 500 ETF\n",
    "    \n",
    "    \n",
    "def handle_data(context, data):\n",
    "    order(symbol('SPY'), 10)  # Order 10 shares every chance you get \n",
    "    record(SPY=data.current(symbol('SPY'), 'price'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use Zipline (rather than Quantopian) to perform the algorithm\n",
    "perf = zipline.run_algorithm(start=datetime(2017, 1, 5, 0, 0, 0, 0, pytz.utc),\n",
    "                      end=datetime(2018, 3, 1, 0, 0, 0, 0, pytz.utc),\n",
    "                      initialize=initialize,\n",
    "                      capital_base=100000,\n",
    "                      handle_data=handle_data,\n",
    "                      data=panel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SPY</th>\n",
       "      <th>algo_volatility</th>\n",
       "      <th>algorithm_period_return</th>\n",
       "      <th>alpha</th>\n",
       "      <th>benchmark_period_return</th>\n",
       "      <th>benchmark_volatility</th>\n",
       "      <th>beta</th>\n",
       "      <th>capital_used</th>\n",
       "      <th>ending_cash</th>\n",
       "      <th>ending_exposure</th>\n",
       "      <th>...</th>\n",
       "      <th>short_exposure</th>\n",
       "      <th>short_value</th>\n",
       "      <th>shorts_count</th>\n",
       "      <th>sortino</th>\n",
       "      <th>starting_cash</th>\n",
       "      <th>starting_exposure</th>\n",
       "      <th>starting_value</th>\n",
       "      <th>trading_days</th>\n",
       "      <th>transactions</th>\n",
       "      <th>treasury_period_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-01-05 21:00:00+00:00</th>\n",
       "      <td>226.399994</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.000794</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>100000.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100000.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>1</td>\n",
       "      <td>[]</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-06 21:00:00+00:00</th>\n",
       "      <td>227.210007</td>\n",
       "      <td>0.000129</td>\n",
       "      <td>-0.000011</td>\n",
       "      <td>-0.000525</td>\n",
       "      <td>0.002780</td>\n",
       "      <td>0.049077</td>\n",
       "      <td>-0.002621</td>\n",
       "      <td>-2273.24612</td>\n",
       "      <td>97726.75388</td>\n",
       "      <td>2272.10007</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-11.224972</td>\n",
       "      <td>100000.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2</td>\n",
       "      <td>[{'price': 227.3236120035, 'order_id': '693803...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-09 21:00:00+00:00</th>\n",
       "      <td>226.460007</td>\n",
       "      <td>0.000745</td>\n",
       "      <td>-0.000098</td>\n",
       "      <td>-0.007814</td>\n",
       "      <td>-0.000530</td>\n",
       "      <td>0.055263</td>\n",
       "      <td>0.009381</td>\n",
       "      <td>-2265.74237</td>\n",
       "      <td>95461.01151</td>\n",
       "      <td>4529.20014</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-10.290439</td>\n",
       "      <td>97726.75388</td>\n",
       "      <td>2272.10007</td>\n",
       "      <td>2272.10007</td>\n",
       "      <td>3</td>\n",
       "      <td>[{'price': 226.5732370035, 'order_id': '1b60cd...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-10 21:00:00+00:00</th>\n",
       "      <td>226.460007</td>\n",
       "      <td>0.000631</td>\n",
       "      <td>-0.000109</td>\n",
       "      <td>-0.006577</td>\n",
       "      <td>-0.000530</td>\n",
       "      <td>0.045143</td>\n",
       "      <td>0.009486</td>\n",
       "      <td>-2265.74237</td>\n",
       "      <td>93195.26914</td>\n",
       "      <td>6793.80021</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>-9.867518</td>\n",
       "      <td>95461.01151</td>\n",
       "      <td>4529.20014</td>\n",
       "      <td>4529.20014</td>\n",
       "      <td>4</td>\n",
       "      <td>[{'price': 226.5732370035, 'order_id': 'eda004...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-11 21:00:00+00:00</th>\n",
       "      <td>227.100006</td>\n",
       "      <td>0.001574</td>\n",
       "      <td>0.000071</td>\n",
       "      <td>0.000904</td>\n",
       "      <td>0.002295</td>\n",
       "      <td>0.044369</td>\n",
       "      <td>0.023095</td>\n",
       "      <td>-2272.14556</td>\n",
       "      <td>90923.12358</td>\n",
       "      <td>9084.00024</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.753326</td>\n",
       "      <td>93195.26914</td>\n",
       "      <td>6793.80021</td>\n",
       "      <td>6793.80021</td>\n",
       "      <td>5</td>\n",
       "      <td>[{'price': 227.213556003, 'order_id': 'a3744d3...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  SPY  algo_volatility  \\\n",
       "2017-01-05 21:00:00+00:00  226.399994              NaN   \n",
       "2017-01-06 21:00:00+00:00  227.210007         0.000129   \n",
       "2017-01-09 21:00:00+00:00  226.460007         0.000745   \n",
       "2017-01-10 21:00:00+00:00  226.460007         0.000631   \n",
       "2017-01-11 21:00:00+00:00  227.100006         0.001574   \n",
       "\n",
       "                           algorithm_period_return     alpha  \\\n",
       "2017-01-05 21:00:00+00:00                 0.000000       NaN   \n",
       "2017-01-06 21:00:00+00:00                -0.000011 -0.000525   \n",
       "2017-01-09 21:00:00+00:00                -0.000098 -0.007814   \n",
       "2017-01-10 21:00:00+00:00                -0.000109 -0.006577   \n",
       "2017-01-11 21:00:00+00:00                 0.000071  0.000904   \n",
       "\n",
       "                           benchmark_period_return  benchmark_volatility  \\\n",
       "2017-01-05 21:00:00+00:00                -0.000794                   NaN   \n",
       "2017-01-06 21:00:00+00:00                 0.002780              0.049077   \n",
       "2017-01-09 21:00:00+00:00                -0.000530              0.055263   \n",
       "2017-01-10 21:00:00+00:00                -0.000530              0.045143   \n",
       "2017-01-11 21:00:00+00:00                 0.002295              0.044369   \n",
       "\n",
       "                               beta  capital_used   ending_cash  \\\n",
       "2017-01-05 21:00:00+00:00       NaN       0.00000  100000.00000   \n",
       "2017-01-06 21:00:00+00:00 -0.002621   -2273.24612   97726.75388   \n",
       "2017-01-09 21:00:00+00:00  0.009381   -2265.74237   95461.01151   \n",
       "2017-01-10 21:00:00+00:00  0.009486   -2265.74237   93195.26914   \n",
       "2017-01-11 21:00:00+00:00  0.023095   -2272.14556   90923.12358   \n",
       "\n",
       "                           ending_exposure          ...            \\\n",
       "2017-01-05 21:00:00+00:00          0.00000          ...             \n",
       "2017-01-06 21:00:00+00:00       2272.10007          ...             \n",
       "2017-01-09 21:00:00+00:00       4529.20014          ...             \n",
       "2017-01-10 21:00:00+00:00       6793.80021          ...             \n",
       "2017-01-11 21:00:00+00:00       9084.00024          ...             \n",
       "\n",
       "                           short_exposure  short_value  shorts_count  \\\n",
       "2017-01-05 21:00:00+00:00             0.0          0.0             0   \n",
       "2017-01-06 21:00:00+00:00             0.0          0.0             0   \n",
       "2017-01-09 21:00:00+00:00             0.0          0.0             0   \n",
       "2017-01-10 21:00:00+00:00             0.0          0.0             0   \n",
       "2017-01-11 21:00:00+00:00             0.0          0.0             0   \n",
       "\n",
       "                             sortino  starting_cash  starting_exposure  \\\n",
       "2017-01-05 21:00:00+00:00        NaN   100000.00000            0.00000   \n",
       "2017-01-06 21:00:00+00:00 -11.224972   100000.00000            0.00000   \n",
       "2017-01-09 21:00:00+00:00 -10.290439    97726.75388         2272.10007   \n",
       "2017-01-10 21:00:00+00:00  -9.867518    95461.01151         4529.20014   \n",
       "2017-01-11 21:00:00+00:00   5.753326    93195.26914         6793.80021   \n",
       "\n",
       "                           starting_value  trading_days  \\\n",
       "2017-01-05 21:00:00+00:00         0.00000             1   \n",
       "2017-01-06 21:00:00+00:00         0.00000             2   \n",
       "2017-01-09 21:00:00+00:00      2272.10007             3   \n",
       "2017-01-10 21:00:00+00:00      4529.20014             4   \n",
       "2017-01-11 21:00:00+00:00      6793.80021             5   \n",
       "\n",
       "                                                                transactions  \\\n",
       "2017-01-05 21:00:00+00:00                                                 []   \n",
       "2017-01-06 21:00:00+00:00  [{'price': 227.3236120035, 'order_id': '693803...   \n",
       "2017-01-09 21:00:00+00:00  [{'price': 226.5732370035, 'order_id': '1b60cd...   \n",
       "2017-01-10 21:00:00+00:00  [{'price': 226.5732370035, 'order_id': 'eda004...   \n",
       "2017-01-11 21:00:00+00:00  [{'price': 227.213556003, 'order_id': 'a3744d3...   \n",
       "\n",
       "                          treasury_period_return  \n",
       "2017-01-05 21:00:00+00:00                    0.0  \n",
       "2017-01-06 21:00:00+00:00                    0.0  \n",
       "2017-01-09 21:00:00+00:00                    0.0  \n",
       "2017-01-10 21:00:00+00:00                    0.0  \n",
       "2017-01-11 21:00:00+00:00                    0.0  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "perf.portfolio_value.plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <a class=\"anchor\" id=\"tradecustom\"></a> Trading Custom Markets (bitcoin example)\n",
    "\n",
    "Here we will discuss how to load in and customize our data for specific markets, in this case [Bitcoin](https://www.investopedia.com/terms/b/bitcoin.asp). More generally, we will see how to set up a custom trading calendar in order to trade in markets that are not necessarily part of the New York Stock Exchange. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}