The Study Of Beta Anomaly On Shanghai And Shenzhen A Share Market

The low return of high Beta stocks,which we refer to as the Beta anomaly,is one of the most worldwide and most persistent anomalies in capital market.In this paper,we use the sample of Chinese A Stock Market between 2006/12 and 2018/12,which is after the non-tradable shares reform,to do the research of Beta anomaly.First,we estimate each stocks' Beta by run the one-factor regression.Then we test whether there is Beta anomaly in this sample by examine the returns of univariate portfolios sorted on Beta.After verify that there is Beta anomaly,we use bivariate portfolios sorted on Beta and other control variables,and Fama-MacBeth regression,to test whether these control variable can explain the Beta anomaly.We find that after control for the turnover rate,the Beta anomaly is no longer significant,which means that maybe disagreement can explain the Beta anomaly.Meanwhile,the lottery property,illiquidity and idiosyncratic volatility can not explain the Beta anomaly.We also do the robustness check of this finding by constructing a new factor to control for turnover rate.Finally,based on the investor sentiment index,we divide the whole sample into three subsamples:pessimistic,normal and optimistic.Then we test the hypothesis that Beta anomaly is caused by the over-optimistic by investors.We find that there is a little Beta anomaly in pessimistic periods,no Beta anomaly in normal periods and significant Beta anomaly in optimistic periods.We hypothesize that this finding obtains because irrational investors tend to be over-pessimistic or over-optimistic in the pessimistic or optimistic periods,which lead to the disagreement of the market.Since high Beta stocks are more sensitive to the whole market,there are more disagreements in high Beta stocks.Under the short-sale restriction,the more disagreement leads to more overpriced.Also,the difference of the ratio of irrational investors in the market when investors are pessimistic and optimistic leads to the difference of the significance of Beta anomaly in these two periods.By test the turnover rate effect in these three periods,we find evidence to support our hypothesis.We also test the lottery effect in these three periods,and give the reason why lottery demand can not explain the Beta anomaly.