Does Realized Skewness Predict The Cross-section Of Equity Returns?

In the real capital market,the rate of return on assets is not normally distributed in the normal state,and often shows a certain degree of skewness,and the study of the skewness phenomenon has been about forty years.However,most of the research and analysis of financial assets in the financial field are based on the normal distribution of financial assets.So far,many studies have shown that most of the distribution of financial assets income is either positive skewness or negative skewness,and shows the characteristics of high peak and fat tail.This is a serious question and challenge to the traditional capital asset pricing model which only studies the two moment characteristics of the yield of financial assets.Especially in the situation of short development time,imperfect organization,information asymmetry,investment knowledge and lack of risk awareness,the yield of assets is more biased.Therefore,the analysis of stock returns in China's stock market can not be as simple as most people in the second-order moment analysis framework,should break through the second-order moment constraints,from the high-order moment level to analyze and study the distribution characteristics of stock returns.According to the above,this paper focuses on the relationship between the high order moment and the cross section yield of Chinese A-share market through the realized volatility,the realized skewness and the realized kurtosis.In this paper,according to the paper of Amaya(2015),we can calculate the weekly value of realized volatility,realized skewness and realized kurtosis according to the daily high frequency data of the stock price.December This paper selects five-minute high frequency data from January 1,2013 to 31,2017 for five years,and selects all the stocks listed on the Shanghai stock exchange and the Shenzhen stock exchange.This analyzs and studies the relationship between the high order moment of each stock and the cross-section yield of the stock,and at the same time constructs a long-short investment strategy——long the investment portfolio which has the highest realized moment,and short the investment portfolio which has the lowest realized moment.By observing the return of the investment strategy constructed,this paper studies the forecast effect of the high order moment on the future return of the stock.At the same time,the three factor(market,market value,and book-to-market)is regressed with the return of the constructed portfolio,and the intercepts of the Fama-French three factor model are obtained.By observing the significance of the intercept,we can verify whether the effect of higher order moment on cross section yield is independent of the three factor.Finally,to further assess the relation between future returns and realized volatility,realized skewness,and realized kurtosis,we carry out various cross-sectional regressions using the method proposed in Fama and MacBeth(1973).The empirical analysis shows that buying stocks in the lowest realized skewness decile and selling stocks in the highest realized skewness decile generates an average return of 36.59 basis points the following weeks with a t-statistic of 3.91.This result is robust and is not captured by the Fama-French factors.The relation between realized kurtosis and next week's stock return is negative but not always significant.The realized volatility is not good enough to predict cross sectionof stock returns.