Some Results Of Stock Price Forecast And Trading Strategy

With the rapid development of society,people’s economic conditions are getting better and better,and their liquid assets are also increasing day by day.More and more people begin to focus on investment.As a mysterious and attractive project,stock has been hot in recent years.In order to better and more accurately study the stock problem,we need to carefully analyze various indicators and various decisions of the stock,master the relevant laws,and put forward reasonable and feasible suggestions for people’s investment.This paper is divided into two parts.The first part is to study the stock trading strategy by using fuzzy system.The second part uses regression analysis,support vector machine and random forest to predict the stock price.Firstly,this paper uses the fuzzy system theory to transform the technology trading strategy in stock into the membership function in fuzzy mathematics,so as to study the change trend of technology trading rules in dynamic stock price with mathematical language and master the basic rules of stock trading.The experimental results show that when there is only one MA rule,the stock price trend is relatively flat and it is impossible to judge the change of price.However,when we join the BBI rule,the stock price jumps greatly.At this time,according to the change trend of the stock price,we can make corresponding strategic adjustments to maximize the return.Secondly,this paper forecasts the stock price through regression analysis,support vector machine and random forest.At the same time,it also combines them to observe whether the combined model can improve the accuracy in predicting the stock price compared with a single model.We selected five representative stocks in the five sectors of the financial industry,as well as their eight indicators: closing price,rise and fall amount,rise and fall range,turnover rate,trading volume,transaction amount,total market value and circulation market value.We predicted the stock price based on the stock historical trading data of eight years from 2014 to 2021.The results show that the prediction accuracy of the single model of regression analysis,support vector machine and random forest is 97.911%,97.796% and 95.717% respectively.After combining them,the prediction accuracy can reach 99.488%,99.688% and 98.454% respectively.This shows that the combined model can improve the accuracy in predicting the stock price,and also provide a reliable reference value for people’s daily investment.