RBF-AR模型在非线性时间序列预测中的应用

RBF-AR model-based nonlinear time series prediction

研究了RBF-AR模型在非线性时间序列中的建模和预测问题,并把它与其它一些新近提出的模型或方法加以了比较.一种结构化非线性参数优化方法用来辨识此模型.数值实验和比较研究表明采用结构化非线性参数优化方法的RBF-AR模型在预测精度上要大大优于其它一些模型或方法.

This paper investigated nonlinear time series modeling and forecasting problem based on RBF-AR model,and the comparisons between RBF-AR model with other newly developed models.A structured nonlinear parameter optimization method（SNPOM） was applied to estimate the model.It is shown by the simulation tests and comparisons that the performance of RBF-AR estimated by SNPOM is superior to other models.