NSFC“计量建模与经济政策研究”基础科学中心学术交流报告会第五讲
报告题目：Econometrics of games: Identification and estimation of static games of incomplete information with complementary or substitutable actions
报告人：江硕，美国华盛顿大学经济学博士，现任厦门大学经济学院、王亚南经济研究院助理教授，研究方向为：计量经济学理论与应用、实证产业组织。
时间：2022年12月8日，12:30-14:00
地点：厦门大学经济楼A318
摘要：In this talk, I first briefly introduce the econometrics of games and then take my current research as an example. In this paper I study the semiparametric identification and estimation of a class of static games of incomplete information that allow for complementarity or substitutability among player’s own actions. Without assuming parametric distributions for the private information, I characterize simple identified sets for structural parameters under mean restrictions. I provide sufficient conditions for point identification. For point-identified models, I provide simple closed-form estimators for structural parameters that converge at root-n rate and are asymptotically normal. Monte Carlo simulations demonstrates the efficacy of the closed-form estimators. As an empirical illustration, I quantify the extra effect on profit of running two types of stores for supermarket chains in strategic entry. Extensions of the baseline results are discussed.
