Research On Multi Factor Stock Selection Model Based On MAX Anomaly

Asset pricing is one of the core contents of modern financial economics.Since Capital Asset Pricing Model(CAPM)is put forward,research results on asset pricing models have been emerging in academic circles.For example,three factor model,four factor model and five factor model is put forward to explore the formation and change law of asset price and predict the future price of assets,so as to achieve the purpose of obtaining excess return in the financial market.However,as far as the stock market is concerned,there are still many anomalies that cannot be explained by the existing asset pricing model.Such as the MAX anomaly exists significantly in the global stock market,and significant excess returns can still be obtained after adjustment by the existing pricing model.It shows that the existing pricing model can not fully explain the MAX anomaly.First of all,this paper takes all the stocks excluding financial and ST classes in China’s A-share main board market as the research sample,for period from January 2001 to December 2021.By averaging the top five daily returns of individual stocks every month,constructing Max5 factor,and verifies the objective existence of the maximum daily return anomaly through portfolio analysis.Secondly,through Fama-Macbeth cross-sectional regression of the research samples,it is found that Max5 factor has a significant negative correlation with the future return of stocks.After controlling market value factor,value factor,momentum factor,reversal factor,idiosyncratic volatility(IVOL)factor and illiquidity(ILLIQ)factor,the negative correlation is still significant.It is worth noting that IVOL factor and ILLIQ factor will also significantly affect the future return of individual stocks.Finally,based on the research conclusion of the maximum daily return anomaly,a multi factor stock selection model based on Max5,IVOL and ILLIQ is constructed.The comprehensive scores of individual stocks are calculated by equal weight scoring method and IC weighted scoring method.The top 10 stocks with comprehensive scores are selected every month to form a portfolio,which can obtain an excess return of144.47% and 80.89% respectively.Based on the maximum daily return anomaly,this paper studies the multi factor stock selection model,summarizes and improves the existing research,which is helpful to a understand the MAX anomaly of China’s stock market clearly,and has practical significance for improving the effectiveness of the market.In addition,the multi factor stock selection model in this paper can obtain stable excess returns and construct an effective stock investment strategy,which provides a reference for the anomaly of MAX for the continuous innovation of factor pricing model,and is conducive to promoting the development of research in the field of asset pricing.