Research On EMD Algorithm And Its Application In Financial Forecasting

This paper uses Empirical Mode Decomposition(EMD)and Wavelet Neural Network(WNN)to study short-term prediction problems of stock prices,which is an important issue of common concern to investors.First,the Empirical Modal Decomposition method is used to decompose the stock index curve into several Intrinsic Mode Function(IMF)and a Residual(Res);then,the Wavelet Neural Network is used to predict each component obtained by the EMD;finally,the predicted values of each component are added to obtain the predictive model of this paper.In this paper,the daily closing price of the Shanghai Stock Exchange Shanghai Composite Index and the stock of China Southern is used for empirical analysis,the results of empirical analysis show that the relative error of the predicted values is less than 0.7%.To make the model easier to operate,this article improves the model.The function obtained by decomposing EMD is reorganized.It is reorganized into two curves: high-frequency curve and low-frequency curve.The model only needs to fit and predict the two curves with the WNN model.So,the exchange rate of RMB against the US dollar has been empirically studied,and the forecast results have been controlled within 0.5%,and the data that can be predicted has been changed from 8 to 20.he results of the two sets of empirical studies show that the improvement of the model improves the prediction accuracy and prediction time.Therefore,the predictive model of this paper is very simple and convenient,and the prediction accuracy is high,it can provide investors with a certain reference when making investments.