A Laboratory Study of Price Formation and Learning in Auction Markets

拥有私人信息的自利交易者如何获得市场交换的均衡价格？价格形成的问题一直是经济理论的核心。这个问题在实际应用中也至关重要，因为许多政策干预（从就业再培训计划到证券市场的SEC报告要求）旨在影响均衡调整，因此其有效性完全取决于价格形成过程。不幸的是，尽管几个世纪的理论辩论和关键的实际影响，潜在的价格形成过程仍然没有得到很好的理解。该项目进行了四组实验室实验，旨在揭示价格形成过程。一套采用连续双拍卖（CDA）市场制度，这是众所周知的促进异常快速和可靠的价格形成在实验室。第二组实验采用单一调用市场（SCM），在现场不太常见，在实验室效率较低，但在理论上比CDA更容易理解。另一组实验考察了称为多重看涨市场（MCM）和统一价格双重拍卖（UPDA）的中间交易机构。该项目通过解决现有实验室数据中的两个主要缺点，为价格形成过程提供了新的经验证据。首先，几乎所有的市场实验都是重复平稳的（或使用固定的附加移动），因此均衡价格在交易期间保持不变（或同步移动）。这样的实验只提供了对价格形成的单一观察，分布在几个交易周期。所提出的实验包含随机值-从公布的均匀分布中独立提取的每个时期的私人成本和价值参数。其次，虽然价格形成过程明显取决于市场制度，但大多数实验中的制度变化不足以确定市场制度为何重要以及如何重要。拟议的实验包含了广泛的制度差异。数据分析将用于测试在最近的文献中发现的三个有希望的理论方法来理解价格形成。

How do self-interested traders with private information arrive at equilibrium prices for market exchange? This question of price formation has always been central to economic theory. The question is also crucial in practical applications because many policy interventions (ranging from job retraining programs to SEC reporting requirements in securities markets) are intended to influence the adjustment to equilibrium and therefore their effectiveness depends entirely on the price formation process. Unfortunately, the underlying price formation process is still not well understood, despite centuries of theoretical debate and crucial practical implications. This project conducts four sets of laboratory experiments designed to shed new light on the price formation process. One set employs the Continuous Double Auction (CDA) market institution, which is known to promote exceptionally rapid and reliable price formation in the laboratory. A second set of experiments employs the Single Call Market (SCM), less commonly seen in the field and less efficient in the laboratory but better understood theoretically than the CDA. The other sets of experiments examine intermediate trading institutions called the Multiple Call Market (MCM) and the Uniform Price Double Auction (UPDA). This project provides new empirical evidence on price formation process by addressing two major shortcomings in existing laboratory data. First, almost all market experiments are repetitively stationary (or use fixed additive shifts), so equilibrium price remains constant (or shifts in lockstep) across trading periods. Such an experiment provides only a single observation of price formation, spread across several trading periods. The proposed experiments incorporates random values - private cost and value parameters drawn independently each period from announced uniform distributions. Second, although it is clear that the price formation process depends crucially on the market institutions, the institutional variations in most experiments are insufficient to identify why and how market institutions matter. The proposed experiments incorporate wide institutional variations. Data analysis will be used to test the three promising theoretical approaches found in recent literature to understanding price formation.