The Determinants of Price in Online Auctions: More Evidence from Quantile Regression

This study explores how seller reputations affect auction prices, and concludes that earlier findings may be biased due to the misspecification of seller reputation. This paper contributes to the literature by offering significant empirical evidence using Taiwanese Internet auction data. Our study reveals that the influence of seller reputations on auction prices is significant, irrespective of the assumptions of linear and non-linear relationships with price. However, failure to consider the non-linear setting of seller reputation would have led us to overestimate the impact of reputations on prices because marginal returns to an incremental increase in reputation declines rapidly for sellers who have more than 15 scores. In addition, using quantile regression, this study finds evidence of considerable differences in their impact on auction prices dependent on the distribution of price levels.