Skewness Risk Of China Stock Market And The Predictability Of Stock Return

Kraus and Litzenberger(1976)added skewness to the traditional CAPM model to add the third moments to the asset pricing model.Harvey and Siddique(2000a,2000b)introduce skewness risk based in ICAPM and argues that apart from undertaking market risk(which can be measured by variance or volatility),the stock market investors also takes skewness risk.Moreover,the skewness risk and future stock excess return has a negative correlation,that the lower skewness(the higher downside risk of stock return)can predict the future stock excess return in theory。Literature usually focus on the prediction of cross-section individual stock returns in US market,rather than aggregate stock market return and Chinese stock market.The contribution of this article includes:In the first place,this paper focuses on the aggregate stock market return predictability on skewness risk.Secondly,this paper focuses on aggregate market skewness rather than individual stock idiosyncratic risk in the Chinese market.Thirdly,this paper will help in investment and risk management in Chinese stock market in reality concerning its research on the Chinese stock market return predictability.In empirical research,based on the method from Amaya,Christoffersen,Jacobs,and Vasquez(2015),this paper adopts the 5-minite high-frequency data from Jan,2003 to Dec,2014 of Shenzhen index and calculates the realized skewness as the proxy variables of skewness risk.After finishing the realized skewness,this paper use the predictive regression model to examine the predictability of next one-month excess return in stock market.The result shows that β is estimated to be-1.86 and significant in 1%significance level.Realized skewness has excellent out-of-sample predictability,and the predictability of realized skewness brings higher investment return.We give an economic explanation for this predictability that the negative correlation between realized skewness and future stock excess return is conducted via influencing the trading activity.Last,we conduct a series of robustness tests.