Munich Personal RePEc Archive

A hidden Markov model for the detection of pure and mixed strategy play in games

Shachat, Jason and Swarthout, J. Todd and Wei, Lijia (2012): A hidden Markov model for the detection of pure and mixed strategy play in games.

[img]
Preview
PDF
MPRA_paper_39896.pdf

Download (700kB) | Preview

Abstract

We propose a statistical model to assess whether individuals strategically use mixed strategies in repeated games. We formulate a hidden Markov model in which the latent state space contains both pure and mixed strategies, and allows switching between these states. We apply the model to data from an experiment in which human subjects repeatedly play a normal form game against a computer that always follows its part of the unique mixed strategy Nash equilibrium profile. Estimated results show significant mixed strategy play and non-stationary dynamics. We also explore the ability of the model to forecast action choice.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.