Information

We propose a dynamic model for forecasting price in on-line auctions. One of the key features of the model is that it is dynamic in the sense that it operates during the on-going live-auction, which makes it different from previous approaches that only consider static models. Our model shares similarity to previous approaches in that it produces price-forecasts for the auction-end; however, it is more flexi-ble in that it incorporates up-to-date price-information that becomes available at any point during the ongoing auction. Our model is also different with respect to how information about price is incorporated. While one part of the model is based on the more traditional notion of an auction’s price-level, another part incorporates the price-velocity and the price-acceleration. In that sense, it incorporates key fea-tures of a dynamic system such as an online auction. The use of novel methodology, functional data analysis, allows us to measure, and subsequently include, the price’s veloc-ity and acceleration. We illustrate our model on a diverse set of eBay auctions across many different book categories. We find significantly higher prediction accuracy compared to standard approaches. 1