Competitive Risk Analysis of Online Limited Supply Auction

In this paper, we study the online limited supply auction problem, in which a seller has k goods to sell and bidders arrive and depart dynamically. From the perspective of the seller, Lavi and Nisan use the competitive analysis to design algorithmic strategies with minimum relative performance risk. We extend the model to propose a risk reward framework, by introducing the risk tolerance and forecast. For the k=1 problem, a risk reservation price algorithm is present and proved to have a flexible competitive ratio. Moreover, we consider the k>1 case, by deriving the risk pricing algorithm, which is competitive and incentive compatible. It is shown that these two online strategies can be analyzed in terms of their riskiness and the potential benefits for using them.