No abstract

No abstract

A growing number of vendors are using a sequence of online auctions to sell large inventories of identical items. Although bidding strategies and bidder behavior in single auctions have been extensively studied, limited research exists on bidding in sequential auctions. We seek to explain how bidders in such an environment learn from the information, and form and update their willingness to pay (WTP). Using a large data set from an online auction retailer, we analyze the evolution of the bidders' WTP as well as the effect of auction design on bidders' WTP in sequential auctions. We see our study in the context of a longitudinal field experiment, in which we were able to track actions of repeat bidders over an extended period of time. Our results show that bidders' WTP in sequential auctions can be explained from their demand characteristics, their participation experience in previous auctions, outcomes in previous auctions, and auction design parameters. We also observe, characterize, and measure what we call a modified demand reduction effect exhibited across different auctions, over time, by multiunit demand bidders. Our findings are important to enable better auction mechanism design, and more sophisticated bidding tools that explore the rich information environment of sequential auctions.

A multiple-inflation Poisson (MIP) model is put forward for analyzing count data that have multiple inflated values. Analogous to the zero-inflated Poisson model (ZIP; Lambert (1992)), MIP assumes a mixture distribution of Poisson and degenerate distributions, where the probabilities for the inflated values are from a cumulative logit model. We explore the properties of the proposed model, with a detailed treatment given to its maximum likelihood estimation. Moreover, we address variable selection by adopting an L1 regularization scheme. Both simulation experiments and an analysis of a health care data set are provided to illustrate the multiple-inflation Poisson model.

No abstract