Support vector machines-based forecasting in Shanghai and Shenzheng 300 index

Based on the characteristic of the time series of Shanghai and Shenzhen 300 index,proposed a method which based on combination of Principal Component Analysis(PCA) and Support Vector Machine(SVM).Firstly,extract the main information of the factors which influence Shanghai and Shenzhen 300 index through PCA,then SVM to train and forecast.Finally,in order to compare predicted value and true value,we calculate the Mean Square Error(MSE) of the predicted closing price in ten time periods,which is 2.11617.Moreover,the bar chart of predicted value and true value testify that predicted trend is almost accurate.Therefore,it can be proved that using Support Vector Machine to predict Shanghai and Shenzhen 300 index is feasible.