A Bayesian Approach to the Best-Choice Problem
研究了买家在随机收到报价时，如何以最大化买到最低价概率为目标决定是否接受报价，并证明当报价分布通过贝叶斯更新时，最优策略是短视的。
Abstract This article considers a version of the so-called best-choice problem. The buyer receives a random sample of price quotations for a good and desires to buy the good at as low a price as possible. After each price quotation is received, the buyer must decide either to accept the price or not, with the objective of maximizing the probability of buying the good at the lowest price. When the distribution of the price quotation is completely known, the optimal buying policy is myopic. The main purpose of this article is to show that this myopic property holds for some interesting price distributions that are updated in a Bayesian manner as successive prices are received.