Prediction And Analysis Of Shanghai Composite Index Based On Machine Learning Method

Stocks are the product of the market economy and have been around for more than 200 years.Along with the rapid development of China's economy,China's stock market is now booming.The number of companies listed on the Shanghai and Shenzhen stock exchanges is increasing year by year.The operating conditions of listed companies are constantly improving and improving.The focus of China's capital finance is being controlled by the money market.Gradually move to the capital market.Now that stocks have become the "barometer" and "alarm" of the entire social economy,its impact on economic development is immeasurable.Based on the brief description of statistical learning theory and support vector machine(SVM),this paper studies the variable selection and kernel function selection of existing securities price forecasting methods based on support vector machine,focusing on the daily closing price of the above index.The case was analyzed and discussed.Firstly,this paper conducts an adaptive empirical analysis of the machine learning model in A-shares,selects the daily data of the Shanghai Composite as the sample data,determines the input variables,selects the kernel function with good prediction performance,and uses the reconstructed kernel model to close the daily.Index analysis and prediction;Furthermore,the fitting results of the empirical analysis are compared with other models we selected to verify the validity of the reconstructed kernel model.Finally,the model backtesting uses the HMM model to separate the sample intervals in this paper according to the state and select them.Representative bull markets,bear markets and shock cities,respectively,in these three states,the predictions obtained by the above model are fitted,and the prediction accuracy is compared with the above-mentioned full-market prediction accuracy,thereby completing the model backtesting.Finally,the conclusion that the LS-SVM model performs better in the stock market forecasting process is obtained.the model obtained in this paper can effectively predict the change of stock index,help investors understand the changes of stock market,and have certain reference significance for stock investment of ordinary investors.