Forecast Of Stock Market Closing Price Based On EEMD Method

Since the establishment of our stock market for more than 20 years,great achievements have been made and become a core part of China's capital market structure.However,China's stock market has a short period of time,is not mature enough,and is subject to large interference from policies.The price of stocks in the stock market fluctuates sharply and manifests itself as a non-linear,non-stationary strong noise market.The stability of the stock market is closely related to the stability of the country's economy.Researching and grasping the law of price fluctuations in the stock market has important practical and theoretical value for investors,government supervision and decision-making departments and academic circles.Based on the actual data of China's stock market,this paper uses the EEMD method to perform empirical mode decomposition of the daily closing price data of Shanghai and Shenzhen cities,and obtains 11 different frequency components and 1 residual items.The high-frequency component,low-frequency component and trend component of the closing price data series correspond to the market fluctuation items related to the stock market fluctuations,the stock market major events items and the stock market overall trend items.Then from the factors that caused the stock price fluctuations,we analyzed the fundamentals and capital aspects of the Shanghai and Shenzhen stock markets,that is,listed companies,market investors,international capital markets,and the domestic macro-environment and other factors,and came to Shanghai and Shenzhen.The stock price volatility is mainly determined by the following three aspects:1.The impact of low-frequency vibration caused by major national policy shocks in the stock market volatility;2.The stock market volatility is affected by high-frequency vibrations caused by normal fluctuations in the stock price of the entire stock market;The overall trend of the domestic market as a whole goes towards long-term trends.Finally,this paper uses the decomposed and reconstructed sequences and establishes the corresponding EEMD-SVM model.It uses the support vector machine algorithm to separately use the high-frequency component,low-frequency component and trend component of the daily closing price of Shanghai and Shenzhen stocks after EEMD decomposition and reconstruction.In the short-term prediction,the prediction sequence reconstruction obtains a new total prediction sequence,and the error comparison between the prediction sequence obtained by the EEMD-SVM model and the prediction sequence obtained by the SVM model shows that the prediction effect of the EEMD-SVM model is better.