一类双线性时间序列模型基于累积量的辨识  

Identification of a Kind of Bilinear Time Series Model Based on Cumulants

作　　者：余勇[1] 万德钧[1] 

机构地区：[1]东南大学仪器科学与工程系

出　　处：《东南大学学报（自然科学版）》1999年第3期139-143,共5页Journal of Southeast University：Natural Science Edition

基　　金：中国船舶工业总公司科研基金

摘　　要：获得了一类双线性时间序列模型的二、三和四阶累积量所满足的差分方程，它们与线性平稳ＡＲＭＡ模型著名的ＹｕｌｅＷａｌｋｅｒ方程相似，可看作是标准ＹｕｌｅＷａｌｋｅｒ方程的推广．因此它们可用于双线性时间序列模型的辨识，并以一个数值仿真例子表明该方法是可行的．Difference equations of the second, third and fourth order cumulants for a kind of bilinear time series model are obtained, and they are similar to the well known Yule Walker equations for the linear and stationary ARMA models. Therefore these equations can be considered as extensions of the standard Yule Walker equations. Thus they can be used for the identification of bilinear models, and a numerical simulation example shows that the method is feasible.

关 键 词：双线性模型 累积量 辨识 时间序列模型 

分 类 号：O211.61[理学—概率论与数理统计] O231[理学—数学]