Research On Stock Price Prediction Based On Data Mining Technology

Since the birth of the stock,it has been more than 200 years old.The most classical theory in stock forecasting is the effective market hypothesis theory(EMH),that is,newly released public information(including its operating conditions,major adjustments,etc.)about the prospects of listed companies.It can be reflected by the current stock price.If we regard all kinds of information published by listed companies as a kind of value,then the rise and fall of stock prices can be considered as a kind of fluctuation around this value.The financial data disclosed by the financial statements of listed companies reflected the operating conditions of the company and became the most important factor affecting the stock price during the period.We synthesize two major stock analysis methods(technical analysis method and fundamental analysis method),firstly do not consider other factors,simply based on market performance,and make predictions based on historical transaction data;then we model the stock price based on the financial report data disclosed by listed companies.Forecasting will eventually weight the two to obtain more accurate forecasting results.This paper first uses the neural network,support vector machine,random forest and other methods in data mining technology to predict the daily closing price of a month based on the market performance of the stock price(k-line chart),and chooses a better prediction effect.The method then uses the performance of the listed company's financial indicators to model the performance of the stock price to predict the average price of the next month's closing price as a reflection of the company's value in the quarter.Finally,the forecasted average daily closing price for one month and the average closing price for the next month are weighted to obtain the best forecast result.In this paper,the value investment is effectively combined with the nonlinear model,a complete stock prediction model is established,and an empirical study is carried out,which proves that it is meaningful to combine the market performance and the financial indicators.