Internet Auctions With Artificial Adaptive Agents: A Study on Market Design
Internet Auctions With Artificial Adaptive Agents: A Study on Market Design
复制标题
DOI：
10.2139/ssrn.926913
复制
发表时间：
2007-02
期刊：
影响因子：
--
通讯作者：
J. Duffy;M. Utku Ünver
中科院分区：
文献类型：
--
作者：
J. Duffy;M. Utku Ünver
我们开发了一个模型，互联网拍卖的目的是了解如何结束这样的拍卖规则（“硬”或“软”关闭）影响投标行为。我们使用有限自动机建模投标策略，并报告涉及人工投标人使用遗传算法更新策略的模拟结果。我们的模型被证明提供晚或早的出价行为，这取决于拍卖是否有一个硬或软关闭规则，根据经验证据。我们报告我们的模型的其他有趣的属性，并从市场设计的角度提供一些结论。
We develop a model of internet auctions with the aim of understanding how rules for ending such auctions (a "hard"- or "soft"-close) affect bidding behavior. We model bidding strategies using finite automata and report results from simulations involving populations of artificial bidders who update their strategies using a genetic algorithm. Our model is shown to deliver late or early bidding behavior, depending on whether the auction has a hard- or soft-close rule in accordance with the empirical evidence. We report on other interesting properties of our model and offer some conclusions from a market design point of view.