Chaotic time series forecasting using online least squares support vector machine regression

A chaotic time series forecasting method based on online least squares support vector machine (LS-SVM) regression is proposed. The difference between the online LS-SVM and offline support vector machine (SVM) is that the online LS-SVM is still effective for the chaotic system with a variation of the system parameter. Four chaotic time series, namely, Chen's system, Rossler system, Henon map and chaotic electroencephalogram (EEG) signal, are used to evaluate the performance. The results verify the ability of the method in chaotic time series prediction.