Are The Size Factor And Value Factor In The Fama-French Three-factor Model The Pricing Factors Of ICAPM?

A lot of evidence shows that the size factor and value factor in the three factors of Fama-French have strong predictive power in cross section,but these two factors do not show the ability to predict stock market returns in time series.What is the reason? Stivers(2018)found a mechanism based on the US market data to make the size factor and value factor can be used as the ICAPM pricing factor: that is,six portfolios are constructed according to size and book-to-market ratio,and then based on the partial least squares(PLS)regression method to obtain an "optimal" linear combination of the six portfolios(denoted as the PLS factor)).Make the PLS factor the most relevant to future excess returns on the market.He found that the PLS factor has a significant predictive power for future market excess returns,and that the dominant forecasting role is the small size portfolios in the size factor and the value portfolios in the value factor.This article uses the analysis framework of Stivers(2018),based on the stock data of Chinese A-share listed companies from 2000 to 2018,to empirically test whether the PLS factor constructed based on the size factor and value factor is able to predict future excess returns on the A-share market.The empirical results in this paper show that:(1)A similar mechanism does exist in the A-share market so that the size factor and value factor can be used as the ICAPM pricing factor,which is basically consistent with the conclusion of Stivers(2018)based on the US market.(2)This paper tests the mechanism by changing the frequency of the data(monthly data is changed to quarterly data),the number of portfolios(constructing 16 or 25 portfolios based on Size and BM),and investigating the A-share markets in Shanghai and Shenzhen respectively.The empirical results show that the PLS factor has a robust ability to predict future excess returns in the A-share market,and that the PLS factor is all dominated by small size portfolios and value portfolios.(3)Different from the overall market conclusion,in the secondary industry,the ability of the PLS factor to predict future market excess returns is mainly brought by large size portfolios and value portfolios.This conclusion is also in contrast with Stivers(2018).The conclusion of the research on the US stock market is different.The PLS factor of the secondary industry in the US market is still dominated by small size portfolios and value portfolios.(4)Overall,it is found that when faced with the Fama-French three factors,the PLS factor still has a strong explanation for the future market’s excess return.This robust prediction ability of the PLS factor makes the PLS factor in the investor’s investment activities have good applicability.