Scalable Optimal Online Auctions

Abstract

This paper provides a set of tools to compute and implement optimal reserve prices for online auctions.

Identification of Auction Models Using Order Statistics

“…11 We build on Song (2004), which considers online English auctions with unobserved competition. Related to our result, Freyberger and Larsen (2017) studies the problem in the classical setting (i.e., separable and continuous UH) using reserve price and two order statistics of bids; without identifying the value distributions, Coey et al (2021) focuses on identification of the optimal reserve price using two order statistics in a general setting. Our approach uses two consecutive order statistics with an instrument, and achieves point identification under nonseparable and finite UH.…”

Secret Reserve Prices by Uninformed Sellers

“…Within this literature, the closest papers to ours are, first, Olimov (2013) who argues that, in eBay auctions for used tractors, sellers use secret reserve prices to run unsuccessful auctions to learn bidders' willingness to pay and use this information in subsequent resale opportunities. 9 Second, Coey, Larsen, Sweeney, and Waisman (2021) argue that, in the context of online auctions, secret reserve prices allow the seller to observe more bids (first and second highest) and dynamically adjust her reserve price in future auctions. Our approach differs from these papers in two respects: the seller learns about the unobserved component of auction heterogeneity rather than bidder's private valuations (or distribution), and our rationale does not rely on dynamics or resale opportunities.…”

Finite-Sample Average Bid Auction

“…Another popular approach is to directly maxmimize the empirical version of (1). Studies related to this approach include Segal (2003); Prasad (2008); Coey et al (2020). This method can be referred to as the empirical Myerson auction because the price is set as if the empirical distribution is the true distribution F .…”