Nonparametric identification and estimation of the generalized second-price auction

Abstract

In this paper, I establish the identification and present a nonparametric estimator for the incomplete information generalized second-price auction model. I recover the distribution of the bidders' quality-adjusted valuations from the CDF of the adjusted bids, win outcomes, and click-through rates. Through the Monte Carlo simulations, I evaluate the finite-sample performance of the proposed estimator. Additionally, I compare the estimator of the incomplete information model to the estimator of the misspecified model assuming that adjusted bids are the result of the bidding according to the locally envy-free equilibrium of complete information model.