Small updates for the paper

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: nonprobsvy
 Type: Package
 Title: Inference Based on Non-Probability Samples
-Version: 0.2.2
+Version: 0.2.3
 Authors@R:
     c(person(given = "Łukasz",
              family = "Chrostowski",

NEWS.md

@@ -2,6 +2,11 @@ nonprobsvy News and Updates
 
 # nonprobsvy (development version)
 
+# nonprobsvy 0.2.3
+
++ changes to the documentation to meet the JSS requirements
++ documentation polishing
+
 ------------------------------------------------------------------------
 
 # nonprobsvy 0.2.2

R/check_balance.R

@@ -1,4 +1,4 @@
-#' @title Checks the variable balance between the probability and non-probability samples
+#' @title Checks Variable Balance Between the Probability and Non-Probability Samples
 #'
 #' @description
 #' Function compares totals for auxiliary variables specified in the `x` argument for an `object` that

R/control_inference.R

@@ -1,4 +1,4 @@
-#' @title Control parameters for inference
+#' @title Control Parameters for Inference
 #'
 #' @description \code{control_inf} constructs a `list` with all necessary control parameters
 #' for statistical inference.

R/control_outcome.R

@@ -1,4 +1,4 @@
-#' @title Control parameters for outcome model
+#' @title Control Parameters for Outcome Model
 #'
 #' @importFrom stats loess.control
 #' @description \code{control_out} constructs a list with all necessary control parameters

R/control_selection.R

@@ -1,4 +1,4 @@
-#' @title Control parameters for the selection model
+#' @title Control Parameters for the Selection Model
 #'
 #' @description \code{control_sel} constructs a list with all necessary control parameters
 #' for selection model.

R/data.R

@@ -1,4 +1,4 @@
-#' Job vacancy survey
+#' @title Job Vacancy Survey
 #'
 #' @description
 #' This is a subset of the Job Vacancy Survey from Poland (for one quarter).
@@ -29,7 +29,7 @@
 #'
 "jvs"
 
-#' Admin data (non-probability survey)
+#' @title Admin Data (Non-Probability Survey)
 #'
 #' @description
 #' This is a subset of the Central Job Offers Database, a voluntary administrative data set (non-probability sample).

R/method_glm.R

@@ -1,4 +1,4 @@
-#' Mass imputation using the generalized linear model method
+#' @title Mass Imputation Using the Generalized Linear Model Method
 #'
 #' @importFrom stats glm.fit
 #' @importFrom stats update

R/method_nn.R

@@ -1,4 +1,4 @@
-#' Mass imputation using nearest neighbours matching method
+#' @title Mass Imputation Using Nearest Neighbours Matching Method
 #'
 #' @importFrom stats update
 #' @importFrom survey svymean

R/method_npar.R

@@ -1,4 +1,4 @@
-#' Mass imputation using non-parametric model method
+#' @title Mass Imputation Using Non-Parametric Model Method
 #'
 #' @importFrom stats update
 #' @importFrom survey svymean

R/method_pmm.R

@@ -1,5 +1,4 @@
-#' Mass imputation using predictive mean matching method
-#'
+#' @title Mass Imputation Using Predictive Mean Matching Method
 #'
 #' @description
 #' Model for the outcome for the mass imputation estimator. The implementation is currently based on [RANN::nn2] function and thus it uses Euclidean distance for matching units from \eqn{S_A} (non-probability) to \eqn{S_B} (probability) based on predicted values from model \eqn{\boldsymbol{x}_i} based

R/methods.R

@@ -20,7 +20,7 @@ extract.nonprob <- function(object, what=c("all", "mean", "se")) {
 
   return(ext)
 }
-#' @title Extracts estimates from the nonprob class object
+#' @title Extracts Estimates from the Nonprob Class Object
 #' @description Returns a \code{data.frame} of estimated mean(s) or standard error(s)
 #' @param object object of of the \code{nonprob} class
 #' @param what what to extract: all estimates (mean(s), SE(s) and CI(s); \code{"all"}; default), estimated mean(s) (\code{"mean"}) or their standard error(s) (\code{"se"})
@@ -50,7 +50,7 @@ pop_size.nonprob <- function(object) {
   object$pop_size
 }
 
-#' @title Returns population size (estimated or fixed)
+#' @title Returns Population Size (Estimated or Fixed)
 #' @description Returns population size that is assumed to be
 #'\itemize{
 #'  \item{\code{fixed} -- if it is based on the `pop_size` argument,}
@@ -94,7 +94,7 @@ pop_size <- function(object) {
 
 # base R methods ----------------------------------------------------------
 
-#' @title Returns the number of rows in samples
+#' @title Returns the Number of Rows in Samples
 #' @description
 #' Returns information on the number of rows of the probability sample (if provided)
 #' and non-probability sample.
@@ -110,7 +110,7 @@ nobs.nonprob <- function(object,
   c("prob" = object$prob_size, "nonprob" = object$nonprob_size)
 }
 
-#' @title Extracts the inverse probability weights
+#' @title Extracts the Inverse Probability (Propensity Score) Weights
 #' @description A generic function `weights` that returns inverse probability weights (if present)
 #'
 #' @param object a \code{nonprob} class object
@@ -142,7 +142,7 @@ weights.nonprob <- function(object,
   object$ipw_weights
 }
 
-#' @title The update method for the nonprob object with changed arguments or parameters
+#' @title Update Method for the Nonprob Object with Changed Arguments or Parameters
 #' @author Maciej Beręsewicz
 #'
 #' @description
@@ -202,7 +202,7 @@ update.nonprob <- function(object, ..., evaluate=TRUE) {
   }
 }
 
-#' @title Returns confidence intervals for estimated mean
+#' @title Returns Confidence Intervals for Estimated Mean
 #'
 #' @description A generic function that returns the confidence interval
 #' for the estimated mean. If standard errors have not been estimated, the function
@@ -277,7 +277,7 @@ confint.nonprob <- function(object,
 }
 
 
-#' @title Returns coefficients of the underlying models
+#' @title Returns Coefficients of the Underlying Models
 #' @description
 #' Returns a \code{list} of coefficients for the selection and the outcome models
 #'

R/nonprob_documentation.R

@@ -1,4 +1,4 @@
-#' @title Inference with non-probability survey samples
+#' @title Inference with Non-Probability Survey Samples
 #' @author Łukasz Chrostowski, Maciej Beręsewicz, Piotr Chlebicki
 #'
 #' @description \code{nonprob} function provides an access to the various methods for inference based on non-probability surveys (including big data). The function allows to estimate the population mean based on the access to a reference probability sample (via the `survey` package),  as well as totals or means of covariates.
@@ -36,8 +36,8 @@
 #' @param start_selection an optional `vector` with starting values for the parameters of the selection equation
 #' @param start_outcome an optional `vector` with starting values for the parameters of the outcome equation
 #' @param verbose a numerical value (default `TRUE`) whether detailed information on the fitting should be presented
-#' @param se Logical value (default `TRUE`) indicating whether to calculate and return standard error of estimated mean.
-#' @param ... Additional, optional arguments
+#' @param se Logical value (default `TRUE`) indicating whether to calculate and return standard error of estimated mean
+#' @param ... Additional, optional arguments (not yet supported)
 #'
 #' @details Let \eqn{y} be the response variable for which we want to estimate the population mean,
 #' given by \deqn{\mu_{y} = \frac{1}{N} \sum_{i=1}^N y_{i}.} For this purpose we consider data integration

R/plot.R

@@ -1,4 +1,4 @@
-#' @title Plots the estimated mean(s) and their confidence interval(s)
+#' @title Plots the Estimated Mean(s) and Their Confidence Interval(s)
 #' @description
 #' Simple plotting method that compares the estimated mean(s) and CI(s) with the naive (uncorrected) estimates.
 #'

R/prints.R

@@ -1,6 +1,3 @@
-# no print doccumentation
-
-
 # general printing --------------------------------------------------------
 
 #' @method print nonprob
@@ -13,8 +10,6 @@ print.nonprob <- function(x, digits=4,...) {
                                                 "ipw" = "inverse probability weighting",
                                                 "dr" = "doubly robust")))
 
-  ## here we should have more information when the doubly robust estimator is applied
-
   cat(sprintf(" - method: %s\n", x$estimator_method))
   cat(sprintf(" - auxiliary variables source: %s\n", ifelse(!is.null(x$svydesign), "survey", "population")))
   cat(sprintf(" - vars selection: %s\n", tolower(x$control$control_inference$vars_selection)))
@@ -66,11 +61,9 @@ print.nonprob <- function(x, digits=4,...) {
     cat(" - caution: since `pop_size' is provided, the variance is estimated assuming that `pop_size' is known (not estimated). If this assumption is wrong, specify the appropriate `var_method' in the `control_inf()` function.\n")
   }
 
-  #cat("Estimated population mean with overall std.err and confidence interval:\n\n")
-  #print(cbind(mean = x$output$mean, SE = x$output$SE, x$confidence_interval))
   invisible(x)
 }
-#' @title Print method for the nonprob_summary object
+#' @title Print Method for the Nonprob_summary Object
 #'
 #' @description
 #' Print method for the `nonprob_summary` object which allows for specification

R/summary.R

@@ -1,4 +1,4 @@
-#' @title Summary statistics for model of the nonprob class
+#' @title Summary Statistics for Models of the Nonprob Class
 #'
 #' @description
 #' Summarises the `nonprob` class object. The summary depends on the type of

inst/WORDLIST

@@ -67,3 +67,4 @@ svydesign
 ’Dplyr’
 Chrostowski
 chrostowski
+JSS