An Application Study on Prediction and Analysis for Ideal Time Series Based on the SVM Method

The support vector machine (SVM) regression principle and basic ideas based on the statistical learning theory are introduced.This method is used to build forecasting models on the ideal time series from 33-mode Lorenz system,and especially the prediction on nonstationary time series are tested and analyzed.It is shown that the SVM method is available for both stationary series and nonstationary ones,and the results are developmental to prediction of real data.