摘要
文献[1]中,我们用有关鞅的中心极限定理,证明了系统辨识中LS估计的渐近正态性。然而[1]中的条件是苛刻的。本文利用Mcleish的相依变量的中心极限定理改进了[1]的结果。
In Yuan's paper [ 1 ], we have proved the asymptotic normality of least square estimator in system identification using the central limit theorem for martingales. However, the conditions of [ 1] are rather harsh. In this artical, we use Mcleish's dependent central limit theorem to improve the above result.
基金
This work is a part of the project supported by National Natural Science Foundation of China
