An Empirical Analysis Of Multi-factor Quantitative Stock Selection Model In China's A-share Market

Quantitative investment refers to the way through people take a quantitative method to obtain stable incomes in the transaction when computer program issues a sale order.The research and application of quantitative investment are of great significance to China's investment field.Multi-factor stock selection model is one of the most important methods to establish stock pool in investment institutions.Therefore,we analyze the multi-factor stock selection strategy,hoping to find a portfolio that overcomes the market index.In the paper,many factors are included in the model such as fundamental factors,technical factors,macroeconomic factors and investor sentiment.In solving the problem of factor's non-linear correlation,we introduce the kernel function to identify and extract the characteristic of candidate factors and propose the method of KPCA and KPCR.We explain the detailed procedures in the sections below.Though experimental verification in the paper,we take the HS300 constituent stocks and their dozens of indicators as input data.The contribution rate of the first five principal components is 92% in the high dimension feature space.Comparing the results with the traditional principal component analysis and partial least squares regression results,it is found that the result of principal component analysis within kernel function is more obvious.Next,we establish an regression equation as the multi-factor stock selection equation which explain the relationships between stock returns and the dimension reduction of indicators.The results show that the model has good predictive effect on the returns.In the process of testing,the frequency when portfolio outperforms market is 83.65%.Finally,it is concluded that in China's stock market,the kernel principal component regression method can be used to identify the pattern and find the most significant factor that influences the stock returns.That means we can use the KPCR method to build a effective portfolio which has big profits and also performs better than the market.