Theodore L.
Turocy
Department of Economics
Texas A&M University
[back to list of papers]
The limit of any convergent sequence of logit quantal response
equilbria is a Nash equilibrium in a strategic game, and the limit of
any convergent sequence of agent quantal response equilibria is a
sequential equilibrium of an extensive game. Using a logarithmic
transformation of action probabilities, it is practical to compute such
sequences, and thereby compute good approximations to Nash and
sequential equilibria. This paper describes the algorithm to compute
the sequences, and outlines the convergence and selection properties of
the method.
Current version dated November 10, 2006. Available in: [pdf].
The algorithm described in this paper has been implemented in Gambit.