支持向量机方法应用于理想时间序列的预测研究

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

简要介绍了基于统计学习理论的支持向量机方法的基本思想和原理,利用该方法对33模Lorenz系统的理想混沌时间序列建立预测模型,并对在此基础上产生的非平稳时间序列进行预测试验研究。结果表明,支持向量机方法不仅对平稳过程有较好的预报能力,也可以适用于非平稳过程,对实际序列的预测有一定的启发意义。

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.