How Accurately Do Structural Asymmetric First-Price Auction Estimates Represent True Valuations?

Structural asymmetric first-price auction estimation methods have provided numerous empirical studies. However, due to the unobserved nature of underlying valuations, the accuracy of estimates is not feasibly testable with field data, a fact that could inhibit empirical auction market designs and applications based on structural estimates. To address the diverse views regarding estimates' accuracy, we provide an analysis of estimates derived from experimental asymmetric auction data in which researchers observe valuations. We test the null of statistical equivalence between the estimated and true value distributions against the alternative of non-equivalence. When advanced models are used, the Modified Kolmogorov-Smirnov test fails to reject the distributional equivalence, supporting structural asymmetric auction estimations for auction market studies