| The multi-factor model studied in this paper is currently the mainstream investment model in the world,and it is also the key research field in China's quantitative investment.Therefore,the research on multi-factor model has specific guiding significance for investors.The selected stock pool was analyzed and studied using the Fama-Macbeth test and the Fama-French model.The empirical ideas of this paper are as follows: Firstly,the author summarizes the general description of the factors of the previous scholars,and summarizes the four major investment strategies.Based on the four major investment strategies,the factors are divided into seven categories,including profitability,valuation factors,The solvency,operational capacity,growth factors,flow factors and other impact factors are used to screen out factors that may meet the screening criteria.Secondly,the above seven types of factors are tested.First,a single factor test is performed.Here we use Is the Fama-Macbeth test,which excludes factors that are not eligible,ie,insignificant factors;finally,multi-factor correlation tests are performed on those factors that pass the verification,and factors with severe collinearity are excluded;Perform multi-factor testing,eliminate those factors with strong collinearity,and use stepwise regression to test the correlation between each factor and the rate of return.In order to more intuitively see the correlation between factor and rate of return.A hierarchical approach is obtained to obtain a complete multivariate stock selection model regression equation.Finally,using the four quarterly data of 2017 as the off-sample data to backtest and test the model,four test periods and four investment portfolios were obtained.In order to maximize the revenue,the investment portfolio containing 20 stocks wasselected.These 20 stocks were given weights,and the conclusions of this paper were drawn.The relevant suggestions were elaborated from the aspects of model improvement and prospect. |
PDF Full Text Request