Epistemic Game Theory: Beliefs and Types
Modelling what each agent believes about her opponents, what she believes her opponents believe about her, and so on, plays a prominent role in game theory and its applications. This article describes Harsanyi’s formalism of type spaces, which provides a simple, elegant representation of probabilistic belief hierarchies. A special emphasis is placed on the construction of rich type spaces, which can generate all ‘reasonable’ belief hierarchies in a given game. Recent developments, employing richer representation of beliefs, are also considered.
KeywordsBelief hierarchies Common knowledge Epistemic game theory: beliefs and types Harsanyi, J.C. Kolmogorov’s extension theorem Monotonicity Polish spaces Preferences Recursive preferences Type spaces Universal type space
- Armbruster, W., and W. Böge. 1979. Bayesian game theory. In Game theory and related topics, ed. O. Moeschlin and D. Pallaschke. Amsterdam: North-Holland.Google Scholar
- Brandenburger, A., Friedenberg, A. and Keisler, H.J. 2006. Admissibility in games. Unpublished, Stern School of Business, New York University.Google Scholar
- Di Tillio, A. 2006. Subjective expected utility in games. Working Paper No. 311, IGIER, Università Bocconi.Google Scholar
- Harsanyi, J. 1967–8. Games of incomplete information played by Bayesian players. Parts I, II, III. Management Science 14:159–182, 320–334, 486–502.Google Scholar
- Morris, S., and H. Shin. 2003. Global games: Theory and applications. In Advances in economics and econometrics (Proceedings of the eight world congress of the econometric society), ed. M. Dewatripont, L. Hansen, and S. Turnovsky. Cambridge: Cambridge University Press.Google Scholar