混沌时间序列的LSSVM预测方法

Prediction Method of the Chaotic Time Series Using Least Squares Support Vector Machine

结合相空间重构理论,提出运用最小二乘支持向量机(LSSVM)建立混沌时间序列的预测模型,并用粒子群优化(PSO)解决LSSVM参数寻优的问题.通过与RBF神经网络构建的预测模型相比较,计算预测模型的均方根误差来评价模型的性能.结果表明:采用PSO优化的LSSVM预测模型的预测精度更高.

Based on the phase space reconstruction theory, prediction model of chaotic time series using least squares support vector machine （LSSVM） was presented in this paper, and particle swarm optimization （PSO） was used to solve the LSSVM parameter optimization problems. By comparing with the established prediction model of RBF neural network, the root mean square error of the prediction model was calculated to evaluate the performance of the model. The resuits show that the PSO optimized LSSVM prediction model has higher prediction accuracy.