Research On Multi-factor Model Optimization Of Quantitative Investment Stock Selection Based On Rsrs Timing Index

With the gradual opening of China's financial market and the continuous improvement of financial order,quantitative investment has been paid more and more attention in China.Quantitative investment refers to the quantitative investment aimed at obtaining stable profits and excess returns by looking for the relationship between economic and financial data.In quantitative investment,stock multi-factor model is a widely used model structure.Its development process starts from single-factor model of CAPM model to APT model and then to the famous three-factor model,which reflects the strong vitality behind the stock multi-factor model.As the most basic work of stock multi-factor model,single-factor mining is naturally the most important.Without significant profitability and robustness factors,the excess return of multi-factor model can not be said.Therefore,this paper takes Shanghai and Shenzhen 300 component stocks as the sample stock pool,chooses the sample stock price data from January 2012 to January 2019,uses the third-party platform to gather the research module based on Python to achieve data cleaning,de-extremum,standardization and market value industry neutralization,and then uses RLM regression model to regression analysis the candidate factor concentration factor,in the candidate factor set.In order to narrow the set of significant factors,the correlation coefficient matrix is obtained by testing the T value of return series,IC mean and IR value.In the long run,the significant factors are BP factor,ROE factor,inversion factor and market value factor.After the financial logic judgment of the effective factors,BP factor and ROE factor are selected as the core factors of the basic stock multi-factor model,ICIR value is used as its corresponding weight value,scoring and sorting method is used as its basic logical framework,the performance of the model is evaluated in the sample per:iod,and RSRS technology timing index is used to optimize the model,so that the model can obtain stable excess.At the same time,reduce the loss of absolute income.After comparing the quantitative performance indicators of the same strategy,we find that the performance indicators of the stock multi-factor model optimized by RSRS significantly outperform the quantitative median indicators of private equity.