Stock Selection Strategy and Model Test in A-share Market

Abstract

Fama and French put forward the famous three-factor model in 1992, book-to-market ratio and market capitalization are added to the market combination factor of CAPM single factor mode to explain the value effect and scale effect, and then proposed the five-factor model in 2015. Based on this background, this paper wants to do regression experiments on these models in A-shares to validate their explanatory ability on stock returns. All the data include the actual data of all non-ST stocks in the A-share market from January 2006 to December 2019, the book-to-market ratio data of all listed companies in the A-share market (non-ST), the return on equity data of all listed companies in the A-share market (non-ST), and the change rate of total assets of all listed companies in the A-share market (non-ST). Using the three-factor model and five-factor model to explain China's A-share market, it is found that these two-factor models have a relatively poor ability to explain the A-share market, and the R-square fitting fluctuates greatly. The three-factor interpretation is better, the R-square fitting has a small fluctuation, and the five-factor R-square fitting has the smallest fluctuation and the largest fit. CAPM single factor model has been proposed for a long time, but the factor model still can explain, and it is still very good. This is because alpha is almost always insignificant. However, the fitting fluctuates greatly and the explanatory power is inferior to Fama-French three-factor model and the Fama-French five-factor model.