摘要
考虑线性回归模型y=x^Tβ+e_1其中误差e是函数系数自回归(FCA)过程.本文研究该模型未知参数的Huber-Dutter估计的渐近性质,在合理的条件下,证明了这些估计量以n^-(1/2)速度渐近于正态分布.
Consider the following linear regression model y=x^Tβ+e, where the error e is a functional coefficient autoregressive (FCA) processes. In this paper we investigate asymptotic behavior of Huber-Dutter (HD) estimators for unknown parameters in the above model. Under some regular conditions, it is shown that the HD estimators are asymptotically normal with convergence rate n^-2^-1 by the using of martingale difference technique.
出处
《应用数学学报》
CSCD
北大核心
2013年第5期881-899,共19页
Acta Mathematicae Applicatae Sinica
基金
国家自然科学基金(No.11071022)
湖北省教育厅(No.Q20122202)资助项目
关键词
线性回归模型
函数系数自回归过程
HD估计
正态分布
linear regression model
functional coefficient autoregressive process
Huber-Dutter estimators
normal distribution