我国股票市场长期异常回报率的计算和检测方法研究

Long-Run Abnormal Returns Calculation and Test in Chinese Stock Market

本文分析了中国学者在经验研究中采用的传统长期异常回报率计算和检测方法,即利用市场平均计算长期异常回报率,利用t检验检测的方法,指出了缺陷,介绍了新的长期异常回报率计算方法和检测方法,说明了改进的机理。改进的长期异常回报率计算方法是基于市值的参考组合法,改进的检测方法是偏度矫正的t检验法和秩和法。采用反复抽样的方法,对新老方法结合中国资本市场数据进行了比较,结果显示了传统方法的局限性以及新方法的优越性。

The article analyses the method used by Chinese researcher in calculating and testing long-run abnormal returns （calculating long-run abnormal returns by market portfolio, testing by standard t test）, point out its deficiency, gives new method to overcome it. The new method uses reference portfolio formed on the base of corporation market value to calculate long-run abnormal returns, uses adjusted-t test or rank test to test abnormal returns. By using these methods to Chinese stock market, the article illustrates the bad performance of old method and discusses its potential damage to empirical result. The empirical result also proves the great improvement of new method comparing to old method.