strategies.FRD2012 {stratEst}

strategies.FRD2012

Description

List of eleven prisoner's dilemma strategies (Fudenberg, Rand, and Dreber 2012).

Usage

data(strategies.FRD2012)

Format

Each strategy is encoded as a data.frame object. The rows of the data frame represent the states of the automaton. The first row is the start state of the automaton. Each data.frame object contains the following variables:

prob.d

Probability to defect.

prob.c

Probability to cooperate.

tremble

Probability of a tremble.

tr(cc)

State transition for the input cc.

tr(cd)

State transition for the input cd.

tr(dc)

State transition for the input dc.

tr(dd)

State transition for the input dd.

Details

c("ALLC","TFT","TF2T","TF3T","T2FT","T2F2T","GRIM","GRIM2","GRIM3","ALLD","DTFT")

The prisoner's dilemma strategies are:

ALLC

Strategy which always cooperates.

TFT

Strategy which cooperates unless the partner defected in the last round.

TF2T

Strategy which cooperates unless the partner defected in the last two rounds.

TF3T

Strategy which cooperates unless the partner defected in the last three rounds.

T2FT

Strategy which cooperates unless the partner defected in either of the last two rounds.

T2F2T

Strategy which cooperates unless the partner defected for two consecutive rounds of the last three rounds.

GRIM

Strategy which cooperates until one player defects, then GRIM defects forever.

GRIM2

Strategy which cooperates until two consecutive rounds occur in which one player defected, then GRIM2 defects forever.

GRIM3

Strategy which cooperates until three consecutive rounds occur in which one player defected, then GRIM3 defects forever.

ALLD

Strategy which always defects.

DTFT

Strategy which starts with defection, then plays according to TFT.

References

Fudenberg D, Rand DG, Dreber A (2012). "Slow to Anger and Fast to Forgive: Cooperation in an Uncertain World." American Economic Review, 102(2), 720-749.

Examples

strategies <- strategies.FRD2012[c("ALLC","ALLD","TFT","GRIM","PTFT")]