基于支持向量机的复杂时间序列预测研究

Research on Complicated Time Series Prediction Based on Support Vector Machines

介绍了相空间重构和基于支持向量机的时间序列预测建模技术,提出了复杂时间序列的多尺度分解方法,对支持向量机回归与预测的各项参数设置进行了试验分析。对股票数据进行建模和预测,结果表明支持向量机对复杂时间序列具有较好的预测效果。

The paper first introduces the technology of phase construction and modeling of time series prediction based on SVM（support vector machines）. Then it proposes, the multiple-scaled decomposing method of complicated time series and analyzes the parameter sensitivity of SVM regression. Finally, it establishes prediction model and applies it to the stock data. Experimental result indicates that SVM is an effective method for complicated time series prediction.