StratModel-class: Strategic Discrete Choice Models

In jrnold/r-strat: Statistical Strategic Models

Description

Objects of class StratModel describe statistical strategic discrete choice models. This class is extended by classes specific to different game forms. The following extended classes are defined

Usage

## Methods with "surprising" arguments
## S4 method for signature 'StratModel'
prAction(object, b)

Objects from the Class

Objects of this class are used internally by the strat function. The functions to estimate the maximum likelihood of a strategic discrete choice model all operate on objects of this class.

Classes derived from StratModel are used to differentiate the specific game forms that strat can estimate. Different game trees only differ in their log-likelihood functions in the estimation of the probabilities of each action. Then adding a new extensive game form is a matter of defining a new class that contains StatModel and an associated prAction method

Slots

The class "StratModel" represents a statistical discrete choice model and contains the slots:

Object of class "character". The names of the unique players. The first element in this vector is the label for player 1, the second element is the label for player 2, ....

Object of class "character". The names of the unique outcome nodes. This is the same vector as colnames(object@Y).

The StratModel class is extended by several classes for specific game forms.

The class StratModel_1-2 has the following restrictions

paths:

Takes the value list(1, c(2,3), c(2,4)).

terminalNodes:

There are three terminal nodes.

utilities:

There are two players.

Methods

prAction

signature(object = "StratModel", b="list") Given coefficients in b, estimates the probability for each action in the game.

jrnold/r-strat documentation built on May 20, 2019, 1:05 a.m.