Integrating kernel principal component analysis with least squares support vector machines for time series forecasting problems

This paper discusses least squares support vector machines (LSSVM) in the time series forecasting problem. Kernel principal component analysis (KPCA) is proposed to calculate principal component. Least squares support vector machines are applied to predict time series. Experimental results show that the performance of LSSVM with feature extraction using KPCA is much better than that without feature extraction. In comparison with PCA, there is also superior Derforrnance in KPCA.