You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Our current payoffs are 0, 1, 2.1, 3. We want to know what happens when these numbers are different. Since the actual numbers don’t matter, only the ratios between numbers, all of the relevant variation can be seen by varying the middle two numbers. In particular: pick two values, b and c, both between 0 and 1. Then:
Instead of “1”, let the payoff of “dove when your opponent plays hawk” be 3b.
Instead of “2.1”, let the payoff of “dove when your opponent plays dove” be 3b + c(3 - 3b). [this represent going “c of the way up the interval” between 3b and 3.]
E.d., b = ⅓ and c = ½ represent the payoff structure 0, 1, 2, 3.
If it’s too much to have two variables, we could just keep “1” and let our single variable be c, again between 0 and 1, where we define the payoff of “dove when your opponent plays dove” as 1 + 2c.
NOTE: This is replacing what I had previously listed as this issue, under lower-priority possibilities: Varying the payoffs of the game, in particular letting the sum of (D, H) + (H, D) > 2(D, D) [more of a robustness check than anything else
see also #44 (one of these is probably a duplicate)
The text was updated successfully, but these errors were encountered:
[MODIFIED ON 11/22]
Our current payoffs are 0, 1, 2.1, 3. We want to know what happens when these numbers are different. Since the actual numbers don’t matter, only the ratios between numbers, all of the relevant variation can be seen by varying the middle two numbers. In particular: pick two values, b and c, both between 0 and 1. Then:
Instead of “1”, let the payoff of “dove when your opponent plays hawk” be 3b.
Instead of “2.1”, let the payoff of “dove when your opponent plays dove” be 3b + c(3 - 3b). [this represent going “c of the way up the interval” between 3b and 3.]
E.d., b = ⅓ and c = ½ represent the payoff structure 0, 1, 2, 3.
If it’s too much to have two variables, we could just keep “1” and let our single variable be c, again between 0 and 1, where we define the payoff of “dove when your opponent plays dove” as 1 + 2c.
NOTE: This is replacing what I had previously listed as this issue, under lower-priority possibilities: Varying the payoffs of the game, in particular letting the sum of (D, H) + (H, D) > 2(D, D) [more of a robustness check than anything else
see also #44 (one of these is probably a duplicate)
The text was updated successfully, but these errors were encountered: