Quantitative Stock Picking Strategy Based On Multi-Factor Model

Since the establishment of the Shanghai and Shenzhen stock exchanges,the Chinese capital market has achieved rapid development,financial products have been continuously enriched and improved,and the supporting regulatory system has also been continuously upgraded.All of these provide a good development environment for the development of quantitative investment in the country.Quantitative investment refers to an investment method that uses a mathematical statistical model introduced in advance and combines the precise operation of a computer to back-test historical data,and attempts to explore the law to obtain stable returns and excess returns.Quantitative investment technology began to rise abroad in the 1950 s,and has developed rapidly in recent years,but it is still in the primary stage of development in China.This article aims to study the more classic multi-factor quantitative stock selection model in quantitative investment,trying to verify the effectiveness of the model for my country's A-share market.First,it introduces some basic concepts,characteristics and development process of quantitative multi-factor stock selection,so that readers have a preliminary understanding of quantification;second,this article will start from the fundamentals and divide the factors that affect the stock price of A-share listed companies into four Categories,and carry out certain cleaning and lag processing on the transaction data available in the capital market from January 2014 to December2018 and the real-time financial data disclosed by the target company,that is,the real financial available at the time The data is used as a basis,and then in strict accordance with the screening steps of effective factors,four effective factors are selected:price-earnings ratio,price-to-book ratio,net profit growth rate,and 60-day average turnover rate;again,we deduplicate the factors and set The effective threshold is 0.5.The results show that the four effective factors obtained in this paper have passed the redundancy test,that is,four effective and non-redundant factors of net profit growth rate,price-earnings ratio,turnover rate and price-to-book ratio are obtained;finally,the four Factors constructed a multi-factor model and three single-factor models,and selected June 2018 to June 2019 as the out-of-sample backtest period to verify the model.Through the importance indicators such as beta,Sharp ratio,total return rate,annualized compound return rate,probability of obtaining positive return,etc.,the risk and return synthesis of the model is investigated,and finally the conclusion is drawn: based on equal weight scoring The target stock selected by the Z-score multi-factor model does have certain advantages over the CSI 300.We found that the multi-factor model can make investors more rational in stock selection and investment,enhance confidence in holding shares,and reduce the impact of emotional noise in the market on stock prices,which has a great positive effect on the development of quantitative investment.Of course,many parts of this article still need to be improved.For example,we can incorporate some macro factors to improve the model based on the micro factors we studied.Now China's A-share market is more tolerant and open,and its linkage with the external market has been improved to a certain extent.The fluctuation of domestic stock prices is sometimes closely related to the international environment,and the weight of macro factors in the model must be increased accordingly.Secondly,the sample backtest interval can be extended appropriately to cover a complete cycle of the market as much as possible,and the performance of the model at each time period should be analyzed.The conclusions drawn from this will be more rigorous.