Research On Quantitative Stock Selection Strategy Based On Data Mining

In recent years,due to the continuous development of the stock market,more and more attention is paid to the quantitative investment technology.China's quantitative investment system is becoming mature gradually.With the continuous improvement of the stock market rules,the number of listed stocks and their associated data are increasing.There is a lot of complex stock data containing useful information,which cannot be found through conventional methods.However,the data mining technology developed in recent years can help us mining data information from the vast number of stock data.By analyzing these data,we can get the information we want.This paper mainly discusses the quantitative stock selection model based on data mining.First of all,more than 3 thousand shares of A stock in the 2013-2015 Shanghai and Shenzhen stock markets are filtered,according to the two conditions:one is,for 3 consecutive years,net assets yield stable and not less than 10%,excluding ST and other stock companies;another is the main business growth rate and net profit growth rate are basically the same and above 10%.After being screened,51 stocks with good fundamentals are retained.Secondl,we selected 17 important indicators which can reflect the profitability,solvency and growth of financial data as the basis of data analysis.Considering the overlap and correlation among the factors,and if there are too many explanatory variables,it will not be easy to distinguish the primary and secondary relations of these indicators,therefore,through the principal component analysis,while retaining the most information of the original data,we select the five comprehensive indicators without correlation,and then achieve the effect of dimensionality reduction.In many data mining algorithms,clustering analysis is particularly easy to understand and has proven to be effective measures in stock selection.In this paper we choose the K-means clustering to study stock selection strategy.Thus the problem of stock selection is changed into the problem of class selection.By comparing the value of K,The optimal K is obtained when K is 5.Therefore,we selected 7 stocks as the final result.With the help of the wind platform,the K lines of history of the selected stocks are viewed,the overall trend of the selected stocks can outperform almost,and they also have upward trend in the future.This proved that the work we have done in this paper has a certain reference function for stock investors.