Design And Application Of Stock Quantitative Investment Strategy Based On Multi-factor Model

The establishment of the Transformable Hedge Partnership Fund,a quantitative fund based on quantitative trading,which was launched in the 1960 s,marked the debut of quantitative investment methods in the investment field.Quantitative investment methods are still relatively young in human investment activities,but due to their inherent characteristics such as discipline,systemicity and accuracy,and driven by the rapid development of modern technologies such as artificial intelligence,big data and cloud computing,the quantitative investment methods have developed tremendously fast.According to reports,the scale of asset management of quantitative hedge funds in the U.S.market in 2018 accounted for about one third of the total scale of hedge fund industry management;as of May 2019,there were 362 quantitative fund products in China,with a total size of about 120 billion yuan.Quantitative fund management has become one of the most important investment tools for global asset management companies.Its underlying products include stocks,currencies,bonds,commodities and various financial derivatives.In order to obtain excess return,various types of quantitative trading strategies have been studied by investment funds companies,each of which has its advantages.However,each strategy has problems such as limited scope of adaptation and better back-testing effects than actual operating effects.Therefore,the development of quantitative investment strategies is a very crucial link of quantitative fund's investment.As an attempt to explore and test the practicability of quantitative investment strategies in China's capital market,this paper builds a multifactor quantitative model based on the scoring method,and improves and optimizes it to a certain extent on the basis of the model.The research in this paper has certain theoretical significance and practical value for the development of quantitative investment strategies.This article uses the Shanghai and Shenzhen 300 index stocks as the research object.The sample research period is from January 2010 to December 2016 for a total of 84 months.The model back test period is from January 2017 to September 2019 for a total of 33 months.Establish a multi-factor quantitative model based on the scoring method and the equal weights allocation of effective factors.Use this basic comprehensive scoring model to score each stock,and use the top 5% of stocks as the portfolio of securities that investors will invest in the next month.The backtesting results show that the strategy's returns can stably exceed the market and obtain excess returns,but the strategy's Sharp ratio,Sortino ratio,information ratio and other indicators that indicate risk-benefit ratio are not high.In order to further study,this article has made more attempts on the weighting method of the large class of agent factors,found that the comprehensive scoring model that uses the maximizes IR of the agent factors to perform the weighting has the best performance.The corresponding indicators of the basic strategy have been improved,which can stably outperform the benchmark return rate,and can well curb downside risks.We examine the performance of the strategy when facing an uncertain market during the epidemic,and finds that the optimized strategy can better resist market risks and steadily outperform benchmarks,This result has some inspiration for investors to use quantitative stock selection strategies to invest.Finally,based on the constructed stock selection strategy and the optimized stock selection strategy,this article gives several thoughts and several ideas for subsequent research.