Prior-Independent Optimal Auctions
研究了卖家不知道买家估值分布时，如何设计拍卖机制以最大化最坏情况下的收益比例，发现二价拍卖在单调递增风险率分布类中是最优的，能保证至少71.53%的基准收益。
Auctions are widely used in practice. Although auctions are also extensively studied in the literature, most of the developments rely on the significant common prior assumption. We study the design of optimal prior-independent selling mechanisms: buyers do not have any information about their competitors, and the seller does not know the distribution of values but only knows a general class to which it belongs. Anchored on the canonical model of buyers with independent and identically distributed values, we analyze a competitive ratio objective in which the seller attempts to optimize the worst-case fraction of revenues garnered compared with those of an oracle with knowledge of the distribution. We characterize properties of optimal mechanisms and in turn establish fundamental impossibility results through upper bounds on the maximin ratio. By also deriving lower bounds on the maximin ratio, we are able to crisply characterize the optimal performance for a spectrum of families of distributions. In particular, our results imply that a second price auction is an optimal mechanism when the seller only knows that the distribution of buyers has a monotone nondecreasing hazard rate, and it guarantees at least 71.53% of oracle revenues against any distribution within this class. Furthermore, a second price auction is near optimal when the class of admissible distributions is that of those with nondecreasing virtual value function (a.k.a. regular). Under this class, it guarantees a fraction of 50% of oracle revenues, and no mechanism can guarantee more than 55.6%. This paper was accepted by Kalyan Talluri, revenue management and market analytics.