In this study, I examine the "Best Offer" mechanism on eBay by structural analysis. The methods I use in this study are based on the literature on structural auction and "Name your own price" models. Thus, in the fi rst chapter, I provide a survey of this literature.

In the second chapter, I conduct a structural analysis, and propose a two step method to estimate the valuations of buyers who use the "Best Offer" format. The method is computationally attractive and can be implemented both parametrically and nonparametrically. It also has the flexibility to capture potential correlation across buyers' valuations and different selling strategies. I test the performance of the estimator in Monte Carlo experiments, and find that the estimator performs reasonably well.

In the third chapter, I apply the estimation method to the used car market on eBay. I estimate buyers' valuations and the determinants of them. I then run counterfactual experiments to find the optimal selling strategies. I find that the sellers behavior in the market is consistent with revenue maximization.

In the fourth chapter, I propose an alternative approach to estimate the distribution of buyers' valuations based on weak behavioral assumptions. The method does not require the buyers to follow any particular offer strategy; therefore, it captures multiple equilibria that are possible to arise in the environment. Utilizing this feature, I test whether the valuation of a buyer who offers once is on average different from that of a buyer who makes multiple offers. The results suggest that sellers can be better off by following a harder bargaining strategy against the buyers who make multiple offers.