ON THE STATIONARITY AND INVERTIBILITY OF A GENERAL BILINEAR TIME SERIES MODEL

In this paper, some statistical properties of a general bilinear time series model BL (n,m,p,q) are discussed. The sufficient conditions for asymptotic stationarity and invertibility are derived and the expressions of the autocovariance are also given for this model. Results of derivation show that the expressions of autocovariance for a bilinear model are quite different to that of a linear model in the case of q≥ 2 .     

陀螺随机漂移过程的实时跟踪算法研究

本文针对陀螺随机漂移过程的非平稳性,提出了采用时变参数AR(P)模型去拟合该过程,并从代数重组的思想出发.研究了一种时变参数AR(P)模型辨识方法。对陀螺随机漂移过程的跟踪结果表明,该方法具有模型结构简单、阶次低、精度高、速度快等特点。     

陀螺随机漂移误差模型的性态判别

本文运用现代概率统计方法,推导二维和三维条件期望值的计算;并通过计算非线性指数及相关系数来判别陀螺随机漂移误差模型的线性与非线性。