A0339
Title: Nonparametric estimation of sponsored search auctions and impacts of ad quality on search revenue
Authors: Dongwoo Kim - Simon Fraser University (Canada) [presenting]
Pallavi Pal - Stevens Institute of Technology (United States)
Abstract: The aim is to present an empirical model of sponsored search auctions in which advertisers are ranked by bid and ad quality. A new nonparametric estimator is introduced for the advertiser's ad value and its distribution under the \textit{`incomplete information'} assumption. The ad value is characterized by a tractable analytical solution given observed auction parameters. Using Yahoo! search auction data, value distributions are estimated, and the bidding behaviour is studied across product categories. It is found that advertisers shade their bids more when facing less competition. Counterfactual analysis is also conducted to evaluate the impact of score squashing on the auctioneer's revenue. The results show that product-specific score squashing can enhance auctioneer revenue at the expense of advertiser profit and consumer welfare.
