多变量时间序列的邻域预测法

Time Series Prediction by Multivariate Next Neighbor Methods

在非线性动力系统中 ,邻域预测法已经很广泛地应用于单变量时间序列 ,本文将这些方法 ,特别是COM法和LL法 ,推广应用于多变量时间序列。当时间序列很短而同时有多个变量的时间序列可测量时 ,多个变量比单变量在长时间序列中更能提供有效信息从而获得更好的预测结果。

In the non linear dynamics system,next neighbor prediction methods have been successfully applied to univariate time series.This paper generalizes these methods,in particular,center of mass prediction (COM prediction)and local linear prediction(LL prediction),to multivariate time series.The use of multiariate prediction techniques is especially useful when time series are short but several variables have been measured simultaneously.These additional variables can sometimes supply more effective information,thus obtaining better prediction results.