基于核主成分分析与最小二乘支持向量机结合处理时间序列预测问题

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

摘要: 探讨了最小二乘支持向量机时间序列预测的方法,提出了用核主成分分析提取主元,然后用最小二乘支持向量机进行预测．通过实验表明,这种方法得到的效果优于没有特征提取的预测．同时与主成分分析提取特征相比,用核主成分分析效果更好．

Abstract: 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.