Rationalizable Conjectural Equilibrium: Between Nash and Rationalizability
For a steady state to be a Nash equilibrium the agents have to perfectly observe the actions of others. This paper suggests a solution concept for cases where players observe only an imperfect signal of what the others' actions are. The model is enriched by specifying the signal that each player has about the actions taken by the others. The solution, which we call rationalizbale conjectural equilibrium (RCE), is a profile of actions such that each player's action is optimal, given the assumption that it is common knowledge that all players maximize their expected utility given their knowledge. The RCE occupies an intermediary position between Nash equilibrium on one hand and Rationalizability style Bernheim-Pearce on the other hand. The concept is demonstrated by several examples in which it refines the rationalizability concept and still is not equivalent to Nash equilibrium.
Year of publication: |
1991-05
|
---|---|
Authors: | Rubinstein, Ariel ; Wolinsky, Asher |
Institutions: | Center for Mathematical Studies in Economics and Management Science (CMS-EMS), Kellogg Graduate School of Management |
Saved in:
Saved in favorites
Similar items by person
-
Renegotiation-Proof Implementation and Time Preferences
Rubinstein, Ariel, (1989)
-
A Comment on the Logic of 'Agreeing to Disagree' Type Results
Rubinstein, Ariel, (1988)
-
A Remark on Infinitely Repeated Extensive Games
Rubinstein, Ariel, (1992)
- More ...