Thresholding the virtual value: a simple method to increase welfare and lower reserve prices in online auction systems

08/21/2018 ∙ by Thomas Nedelec, et al. ∙ 0

Second price auctions with reserve price are widely used by the main Internet actors because of their incentive compatibility property. We show that once reserve price are learned based on past bidder behavior, this auction is not anymore incentive compatible. Through a functional analytic rather than game theoretic approach, we exhibit shading strategies which lead to large increase of revenue for the bidders. In the symmetric case, we show that there exists a simple equilibrium strategy that enables bidders to get the revenue they would get in a second price auction without reserve price. We then study the consequences of this result on recent work on collusion in second price auctions and prove that the proposed bidding strategies are robust to some approximation error of the auctioneer.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.