基于实验经济学方法的证券市场信息有效性研究

应用实验方法探讨证券市场信息有效性问题是金融研究的一个新方向.50名实验参与人通过浙江大学经济学院开发的实验经济学计算机模拟系统,分别在三组信息结构不同的实验市场中进行虚拟证券交易.实验市场的交易数据表明,在信息完全对称的条件下,交易者并非能够应用所有信息形成一致的理性预期;在有关分红的事前信息不对称条件下,不知情的交易者也未能通过对价格的观察实现信息的有效扩散,价格偏离均衡价格的程度非常显著;而在有关成交价格的事后信息不对称时,信息的有效程度较高,市场价格基本上收敛于理性预期的均衡价格.

An Experimental Research on Information Efficiency of Securities Markets

Information efficiency of securities markets is always an important coomponent of financial theories.Although a great many propositions have studied on this area,how to test them with field data is still a big problem.Information efficiency requires to be measured by the degree of price derived from the fundamental value which cannot be evaluated accurately in the real world.Experimental methods allow us to investigate into propositions on the information efficiency in a controlled environment and evaluate the fundamental values accurately. With reference to researches of experimental economics in the West,we apply the computer experimental system developed by Zhejiang University and recruit participants to build simulation laboratory securities markets.We set three sessions of experiments with different information environment including information symmetry,pre-trade information asymmetry and post-trade information asymmetry.Each session consists of three trials.According to the experimental data,we find that when the laboratory environments are information symmetrical,traders cannot use the information to form common rational expectations.It implies that symmetrical information is not enough for traders to form common expectations and markets cannot realize equilibrium instantaneously.Compared with information symmetrical environments,pre-trade information asymmetry will induce prices to derive from fundamental values more seriously and produce more bubbles,while prices in post-trade information asymmetry markets are very close to the fundamental values.Those experimental results imply that information efficiency is reduced by post-trade transparency and improved by pre-trade transparency.We use trader behavior patterns to explain those results and propose that transparency of efficient information is important to securities regulation.