43 Pages Posted: 7 Mar 2016
Date Written: March 4, 2016
We study the identification and estimation of a first price auction model in which bidders face ambiguity about the valuation distribution and have maxmin expected utility. We exploit variation in the number of bidders to nonparametrically identify the true valuation distribution and the most pessimistic distribution in the bidders’ set of prior distributions. The identification result is extended to allow for constant relative risk aversion and separable unobserved auction heterogeneity that can be correlated with the number of bidders. We propose a flexible Bayesian estimation method based on Bernstein polynomials. Monte Carlo experiments show that our method estimates parameters precisely and chooses the reserve prices with (nearly) maximal revenues – whether there is ambiguity or not. Incorrectly assuming no ambiguity may, however, induce estimation bias, which can lead to a suboptimal reserve price and substantial revenue loss. We apply our method to a sample of U.S. timber auctions and find evidence of ambiguity.
Keywords: first-price auction, identification, ambiguity aversion, maxmin expected utility, Bayesian estimation
JEL Classification: C11, C44, D44
Suggested Citation: Suggested Citation
Aryal, Gaurab and Grundl, Serafin and Kim, Dong-Hyuk and Zhu, Yu, Empirical Relevance of Ambiguity in First-Price Auctions (March 4, 2016). Available at SSRN: https://ssrn.com/abstract=2742704 or http://dx.doi.org/10.2139/ssrn.2742704