| Skewness is not only a major factor in describing the distribution of stock's return,but also closely related to expected return of stock.Highly positively skewed stocks are considered to have lottery characteristics,while its highly negatively skewed counterpart may suffer disaster risk.Therefore,investors prefer the former one and aversion to the latter one.For the sake of obtaining highly positively skewed stocks,investors need to accept lower expected returns.Meanwhile,in order to make investors attracted by highly negatively stocks,such stocks have to provide higher expected returns.The skewness factor not only has its own stock pricing capability,but also is the main cause behind some stock market anomalies.Considering the high proportion of retail investors in China's stock market participants,it is of great significance to study the relationship between retail preferences and skewness premiums.On the basis that skewness is an important pricing factor of China's stock market,this paper explains the "realized kurtosis puzzle" and "market beta anomaly" existing in China's stock market from the perspective of skewness preference.Secondly,considering the heterogeneity of different types of investors5 preferences on skewness,this paper discusses the impact of retail preference on the relationship between skewness and its stock pricing ability.Finally,this paper constructs a new skewness preference measure and empirically demonstrates its stock pricing ability.The main conclusions of this paper are as follows:Firstly,this paper proves the skewness(lottery)preference can effectively explain the "realized kurtosis puzzle" of the China's stock market.The kurtosis corresponds to the thick tails on both sides,high kurtosis corresponds to fat tails on both sides and under risk-aversion assumption investors' dislike of left-tail loss outweighs their preference for right-tail gain.Therefore,high kurtosis characteristic of stock should predict high expected return.However,the high-frequency data based empirical results on China's stock market are just the opposite,which we refer to as the "realized kurtosis puzzle".The empirical research in this paper shows:(1)Using the double sorts and firm-level cross-sectional regression methods,this paper demonstrates investors'skewness preference can well explain the "realized kurtosis puzzle".(2)Our further empirical research verifies stocks with higher retail investors' shareholding proportion and unavailable for short show stronger "realized kurtosis puzzle".(3)The puzzle is time-varying.It is particularly significant in high skewness preference periods while less apparent in low skewness preference times.Secondly,this paper proves that the skewness(lottery)preference can effectively explain the "market beta anomaly" of the China's stock market.CAPM states there should be a positive correlation between stock's market beta and its expected return,but the empirical results of this paper based on China's stock market are contrary to it-low market beta stocks enjoy high returns,while high market beta stocks suffer low returns.This paper examines five possible explanations for '"market beta anomaly"-accidental result under data mining,leverage constraint,agency problem,downside market beta,and skewness preference.The empirical evidences do not support the first four interpretations,but actively support the explanation of skewness preference-the high market beta stock portfolio has the high skewness characteristics preferred by investors,resulting in high market beta corresponding to low return.The empirical research in this paper shows that:(I)Based on the double sorts,firm-level cross-sectional regression,and time series regression of factor mimicking portfolio,'"market beta anomaly" becomes insignificant after controlling the skewness variable.(2)Market beta has certain time variability.In the period of low skewness preference,in accordance with CAPM market beta shows a positive correlation with stock's return;when market performance is poor,the anomaly is no longer significant.(3)The higher the market beta of stock portfolio is,the higher the proportion of stocks with high retail shareholding ratio.Thirdly,the stronger the preference of retail investors,the higher the premium of skewness(lottery).Investors prefer stocks with high skewness characteristics,thus such stocks should have lower expected returns.However,not all investors have the same preference for the high skewness characteristics of stocks.This paper constructs the retail preference indicator by making use of institutional investors' shareholding ratio and a series of stock characteristics preferred by retail investors.Based on this indicator,this paper further studies the relationship between retail preference and skewness premium in China's stock market.The main empirical conclusions of this paper are as follows:(1)The results of double sorts and firm-level cross-sectional regression demonstrate that the strength of retail preference indicators can significantly affect the relationship between skewness and its stock pricing ability.The stronger the retail preference is,the higher the skewnss premium and the stronger its pricing power.(2)Based on the sub-sample analysis method,this paper shows that after excluding certain amount of stocks preferred retail investors,the skewness variables in the remaining stocks lose their pricing power.Stocks with skewness premium account for a small proportion of the whole market value.(3)As retail investors are sensitive to market performance fluctuation,skewness premiums driven by retail investors are more pronounced during periods with good stock market performance.Fourthly,this paper constructs a new stock skewness indicator,relative signed variation(RSV),and tests its stock pricing capability in China's stock market."Good"uncertainty reflects the positive skewness attribute of the stock,while "bad" uncertainty shows the negative skewness characteristics of the stock.With reference to the semi-variance measure proposed by Barndorff-Nielsen et al.(2010),this paper constructs the index of relative sign variation(RSV)as a measure of "good" uncertainty and "bad"uncertainty,and analyze its impact on Chinese stock pricing.Based on the empirical study of 5-minute high-frequency data of Chinese A shares from 2007 to 2017,this paper finds that:(1)There is a negative correlation between RSV and stock returns.This effect is very significant both economically and statistically in single sort,double sorts and firm-level cross-sectional regression.(2)RSV is an important pricing factor in Chinese stock market,and its stock pricing power is stronger than the realized skewness.(3)The impact of RSV on China's stock market is state-dependent.Compared to the state with high economic prosperity,RSV leads to a higher stock risk premium in a state of low economic prosperity.(4)The performance of portfolios constructed based on RSV is significantly better than the market excess returns portfolio,SMB portfolio and HML portfolio. |
PDF Full Text Request