Stock Investment Strategy Based on Decision Tree

Based on the decision tree model, we propose an S-DT investment strategy which predicts the closing price trends of international and domestic stocks. We train the history data of stocks through decision tree model. In order to make the stock price trend prediction accuracy is higher, we choose indexes that are highly correlated with closing price as input variables based on synergy factor, information entropy and grey association rules. Through a large number of empirical data analysis, we found the predictive accuracy of S-DT investment strategy is 89.5%, which is 11.6% higher than that of GRA-DT, ANN, DM and so on.