基于沪深300成份股的多因子量化选股策略研究

Research on Multiple-Factor Quantitative Stock Selection Strategy Based on CSI 300 Stocks

选取沪深300成份股作为样本股,截取2007-2016年财务数据和行情数据,通过模糊C均值聚类算法进行有效但冗余因子的剔除,构建多因子选股模型,将投资组合收益作为检验模型有效性的依据开展研究。结果表明,市盈率、市净率、每股收益等因子对股价波动规律有较强的关联性,多因子选股模型应用于投资实践能够取得稳定的相对于沪深300指数收益的超额收益,模糊C均值聚类算法与多因子选股模型的结合获得了较好的验证,为量化投资研究提供了新思路。

The CSI 300 constituent stocks are selected as the sample stocks and interception of 2007-2016 financial data and market data are used to build multiple-factor stock selection model through fuzzy C-means clustering algorithm eliminating the effective but redundancy factors. This paper makes a research based on the portfolio returns to test the effectiveness of the model. The results show that the p/e ratio,price-to-book ratio,earnings per share and other factors strongly correlate with the stock price fluctuation rule; the multiple-factor stock selection model is applied to the investment practice to obtain stable excess returns relative to the CSI 300 index; the combination of fuzzy C-means clustering algorithm and multiple-factor stock selection model is well verified,which provides a new idea for quantitative investment research.