基于R^2的中国股市私有信息套利分析

Private Information Arbitrage in Chinese Stock Market:A Study Based on R^2

R2在统计学中是模型对实证数据拟合优度的度量指标,自Roll(1988)"R2"开创性研究后,R2在金融学研究中得以更为广泛地度量应用。遗憾的是,在对Roll(1988)进一步研究中所进行的大胆理论延伸,造成了R2在金融领域度量标准在经济意义上的混乱。本文通过对现有文献的梳理、解读和评价,采用逻辑演绎的方法推导出R2可作为私有信息套利的度量指标。本文对2005—2007年上证180中的140只股票进行实证研究表明:(1)几乎所有股票都存在着私有信息套利,且60%的股票私有信息套利程度较深。(2)通过面板模型进一步分析了影响私有信息套利的主要变量,发现上市公司第一大股东持股比例、限售股比例与R2存在显著的负相关关系,机构投资者持股比例、股权分置改革完成时间、公司规模与R2存在正相关关系。

R2 is the proportion of variability in a data set that is accounted for by a statistical model. Roll （1988） has successfully introduced R2 into the analysis of information problem in stock market. Afterwards, Re has been widely applied in various studies on financial market. However, in some cases, Roll＇s R2 is misunderstood and brings controversial results. Our paper has thoroughly reviewed current literatures on R2 and justifies that R2 can be used as a proxy for measuring private information arbitrage activities in the stock market. In this paper, 140 out of SHSE 180 stocks have been analyzed and the results show that （1） most of them have encountered the problem of private information arbitrage. And 60% among these stocks might experienced serious arbitrage problem. （2） Also, our panel analysis on the determinants of private information arbitrage shows that the ownership of prime shareholder, the percentage of non-tradable shares will significantly reduce R2 and hence increase arbitrage opportunity. While institutional investors, stock market reformation, sizes of listed companies will increase R2 and reduce arbitrage opportunity.