Multiple-Factor Stock Selection Model Based On China's A-share Market

As a new way of investing with computer and mathematical methods,there are already very mature systems ofquantitative investment in foreign countries,but the research and application of quantitative investment in domestic market is still very limited.Multi-factor model is one of the most widely used methods in quantitative investment.The construction of multiple-factor model in developed countries has been relatively perfect and arbitrage space is limited,but the multi-factor stock selection model for China's A-share market and lack of perfection still have valuable research significance.Based on the model of multi-factor stock selection,this paper chooses the data of financial indicators and market indicators of all A-share listed companies from January 2006 to March 2017.First,17 candidate factors are selected,than based on the five test indicators,which consist of the absolute value of the information coefficient,the factor annual excess return rate,the excess return hypothesis test P value and the factor information.Then eight effective factors are selected,they are price-earnings ratio,book value,nearly the same month,near the March increase,ROE growth rate expected,the dividend yield and the expected next year's net profit growth rate.All the sample stocks in each period are sorted by assigning a certain weight to the valid factor,and the 50 top stocks are selected for the construction of the portfolio.When the performance of the portfolio is analyzed,it is found that the combination of the time The yield is as high as 56.31%,relative to the million in the A-index excess rate of return has reached 37.51%,and any single factor to build a portfolio of return on income should be high.The use of multiple-factor stock selection model can effectively improve the weak position of individual investors in the capital market,and avoid the misjudgment of some subjective factors,which has a very positive effect on the development of quantitative investment in China's A-share market.However,there are still many shortcomings in this paper,such as the candidate factor and factor validity test indicators is not comprehensive enough.In the future,we will also improve these shortcomings in order to establish a more comprehensive multiple-factor stock selection model and apply it to the real market.