Atlas: Random Belief Equilibrium: Theory and Applications to Experimental Data. by Claudio Mezzetti

Atlas home || Conferences | Abstracts | about Atlas

First World Congress of the Game Theory Society (Games 2000)
July 24-28, 2000
Basque Country University and Fundacion B.B.V.
Bilbao, Spain

Organizers
Ehud Kalai, Federico Valenciano

View Abstracts
Conference Homepage

Random Belief Equilibrium: Theory and Applications to Experimental Data.
by
Claudio Mezzetti
University of North Carolina, Department of Economics
Coauthors: James W. Friedman (University of North Carolina)

We define Random Belief Equilibrium (RBE) in finite, normal form games. We assume that a player's beliefs about the strategy choices of others are randomly drawn from a belief distribution. This distribution is dispersed around a central strategy profile, the focus. At an RBE: (1) Each player chooses a best response relative to her drawn beliefs. (2) The expected choice of each player coincides with the focus of the other players' belief distributions.

We provide a characterization of the limiting RBE as players' beliefs converge to certainty. The limiting RBE is a trembling hand perfect Nash equilibrium, but not all perfect equilibria are limiting RBE, and being a limiting RBE is neither necessary nor sufficient for being weakly proper.

Using maximum likelihood estimation, we find that RBE fits the data from several experimental games significantly better than the Quantal Response Equilibrium of McKelvey and Palfrey.

Date received: June 12, 2000