Finitely additive beliefs and universal type spaces. (English) Zbl 1161.91009

Meier, Martin

Ann. Probab. 34, No. 1, 386-422 (2006).

Summary: The probabilistic type spaces in the sense of J. C. Harsanyi [Manage. Sci., Theory 14, 159–182 (1967; Zbl 0207.51102), 320–334 (1968; Zbl 0177.48402), 486–502 (1968; Zbl 0177.48501)] are the prevalent models used to describe interactive uncertainty. In this paper we examine the existence of a universal type space when beliefs are described by finitely additive probability measures. We find that in the category of all type spaces that satisfy certain measurability conditions (\(\kappa\)-measurability, for some fixed regular cardinal \(\kappa\)), there is a universal type space (i.e., a terminal object) to which every type space can be mapped in a unique beliefs-preserving way. However, by a probabilistic adaption of the elegant sober-drunk example of A. Heifetz and D. Samet [Games Econ. Behav. 22, No. 2, 260–273 (1998; Zbl 0895.90201)] we show that if all subsets of the spaces are required to be measurable, then there is no universal type space.

MSC:

91A40	Other game-theoretic models
91A35	Decision theory for games
28A10	Real- or complex-valued set functions
60A05	Axioms; other general questions in probability

Citations:

Zbl 0207.51102; Zbl 0177.48402; Zbl 0177.48501; Zbl 0895.90201

References:

[1]	Aumann, R. J. and Brandenburger, A. (1995). Epistemic conditions for Nash equilibrium. Econometrica 63 1161–1180. · Zbl 0841.90125
[2]	Battigalli, P. and Siniscalchi, M. (1999). Interactive beliefs and forward induction. Working Paper ECO 99/15, European Univ. Institute. · Zbl 0972.91020
[3]	Brandenburger, A. and Dekel, E. (1993). Hierarchies of beliefs and common knowledge. J. Econom. Theory 59 189–198. · Zbl 0773.90109
[4]	Devlin, K. (1993). The Joy of Sets , 2nd ed. Springer, New York. · Zbl 0792.04001
[5]	Harsanyi, J. C. (1967/68). Games with incomplete information played by Bayesian players, I–III. Management Sci. 14 159–182, 320–334, 486–502. · Zbl 0207.51102
[6]	Heifetz, A. (1993). The Bayesian formulation of incomplete information—the noncompact case. Internat. J. Game Theory 21 329–338. · Zbl 0794.90009
[7]	Heifetz, A. (2002). Limitations of the syntactic approach. In Handbook of Game Theory 3 (R. J. Aumann and S. Hart, eds.) 1682–1684. North-Holland, Amsterdam.
[8]	Heifetz, A. and Mongin, P. (2001). Probability logic for type spaces. Games Econom. Behav. 35 31–53. · Zbl 0978.03017
[9]	Heifetz, A. and Samet, D. (1998). Knowledge spaces with arbitrarily high rank. Games Econom. Behav. 22 260–273. · Zbl 0895.90201
[10]	Heifetz, A. and Samet, D. (1998). Topology-free typology of beliefs. J. Econom. Theory 82 324–341. · Zbl 0921.90156
[11]	Horn, A. and Tarski, A. (1948). Measures in Boolean algebras. Trans. Amer. Math. Soc. 64 467–497. · Zbl 0035.03001
[12]	Łoś, J. and Marczewski, E. (1949). Extensions of measure. Fund. Math. 36 267–276. · Zbl 0039.05202
[13]	Mertens, J. F., Sorin, S. and Zamir, S. (1994). Repeated games. Part A. Background material. CORE Discussion Paper 9420, Univ. Catholique de Louvain. · Zbl 0448.90080
[14]	Mertens, J. F. and Zamir, S. (1985). Formulation of Bayesian analysis for games with incomplete information. Internat. J. Game Theory 14 1–29. · Zbl 0567.90103
[15]	Savage, L. J. (1954, 1972). The Foundations of Statistics . New York, Wiley. [Second edition (1972) Dover, New York.] · Zbl 0276.62006
[16]	Stalnaker, R. (1998). Belief revision in games: Forward and backward induction. Math. Social. Sci. 36 31–56. · Zbl 0963.91016