【主讲】Ritesh Jain (俄亥俄州立大学 博士)
【主题】Rationalizable Implementation of Social Choice Correspondences
【时间】2018年1月25日 (周四) 15：30-17：00
【地点】上海财经大学经济学院楼701室
【语言】英文
【摘要】 I study which social choice rules can be implemented by a mechanism when rationalizability is considered as the appropriate prediction of behavior in games. Introduced by Bernheim (1984) and Pearce (1984), rationalizability is a central solution concept in game theory and makes weaker predictions than Nash equilibrium. The main result of my paper asserts that in a society with more than three agents, a social choice rule can be implemented in rationalizable strategies if it satisfies three properties: r-monotonicity, no worst alternative condition, and Theta_F-distinguishability. Out of these, r-monotonicity is shown to be necessary. As an application of the main result, I show that Pareto correspondence can be implemented in rationalizable strategies. On the other hand, in general, no Condorcet extension can be implemented in rationalizable strategies.
