Are Structural Estimates of Auction Models Reasonable? Evidence from Experimental Data

Recently, economists have developed methods for structural estimation of auction models. Many researchers object to these methods because they find the rationality assumptions used in these models to be implausible. In this paper, we explore whether structural auction models can generate reasonable estimates of bidders' private information. Using bid data from auction experiments, we estimate four alternative structural models of bidding in first-price sealed-bid auctions: 1) risk neutral Bayes-Nash, 2) risk averse Bayes-Nash, 3) a model of learning and 4) a quantal response model of bidding. For each model, we compare the estimated valuations and the valuations assigned to bidders in the experiments. We find that a slight modification of Guerre, Perrigne and Vuong's (2000) procedure for estimating the risk neutral Bayes-Nash model to allow for bidder asymmetries generates quite reasonable estimates of the structural parameters