BAYESIAN LEARNING IN REPEATED GAMES

This paper studies repeated games with a finite number of players, a finite number of actions, discounted payoffs, and perfect recall. The players' initial expectations are given by a common prior distribution over player types, a type being a discount rate and payoff matrix. The main result is that Bayesian Nash equilibria have the property that, with probability one, players' expectations asymptotically approach the set of Nash equilibria for the repeated game specified by the true types. Journal of Economic Literature Classification Number: C72. (C) 1995 Academic Press, Inc.