CPP Quantitative Stock Selection Strategy Design Considering Industry Factors

The difference between quantitative investment and traditional investment is to rely on mathematical models and data to find investment targets,implement investment strategies and seek stable and continuous returns.Different industries have different characteristics,from the perspective of equity investors,stock investment in different industries,for future yields will produce very big effect,and the different characteristics of industries is also important basis to pick stocks,through the analysis of the different characteristics of different industries,on this basis,to pick stocks,improve investment returns.It is very important to correctly understand the importance of industry factors,clarify the mechanism of industry factors on stock price,and systematically explain the industry effect of stock return rate.Although the industry factor is a very key factor in stock selection,the index of industry factor is not easy to quantify,so it is difficult to consider the industry factor in quantitative investment.Based on the perspective of industry risk,this paper attempts to construct industry factors unique to China's market from two aspects of industry concentration degree and industry innovation degree.The constructed industry factors were added into the machine learning model to predict the probability of stock returns,and the results were compared with the multi-factor stock selection model without industry factors.In this paper,industry factors were firstly constructed according to theoretical analysis,single factor detection was carried out for industry factors,and IC value of industry factors was calculated.Secondly,back-measurement method was established on the poly quantitative platform to carry out back-test analysis of multi-classification probability stock selection strategies for industry factors and non-industry factors respectively.The results showedthat the cumulative return of the stock selection strategy without adding industry factors was 455.14%,the cumulative return of the stock selection strategy with adding industry factors was 518.59%,and the benchmark return of the same period was 61.37%.In the real offer simulation,the multi-classification probability prediction of the stock selection strategy with adding industry factors still achieved59% annualized return.After that,this paper combined the multi-category probability prediction stock selection strategy with RSRS(relative strength of resistance support)timing strategy,and found that the strategy could effectively reduce the maximum pullback and increase the cumulative return rate to 684.31% after adding the timing stop loss module.Finally,the paper puts forward investment Suggestions based on the industry perspective of multi-category probability prediction and stock selection strategy.