作者: Stephen G. Donald , Harry J. Paarsch
DOI: 10.2307/2526953
关键词:
摘要: In applications of game theory to auctions, researchers assume that players choose strategies based upon a commonly known distribution the latent characteristics. Rational behaviour, within an assumed class distributions for process, imposes testable restrictions data generating process equilibrium strategies. Unfortunately, support often depends all parameters characteristics, making standard application maximum likelihood estimation procedures inappropriate. We present piecewise pseudo-maximum estimator as well conditions its consistency and asymptotic distribution. empirical (or unobserved) characteristics is common knowledge game. For example, in independent private values model auction, valuations bidders. Moreover, each bidder knows his opponents know valuations, he knows, etc. Based their given realization from valuation distribution, are bids which maximize expected pay-offs winning auction. Given this informational structure, can be characterized by appealing particular concept (e.g., Bayesian-Nash).