On robust estimation in the first order autoregressive processes

被引：7
|
作者：
Haddad, JN [1 ]
机构：
[1] Amer Univ Beirut, Beirut, Lebanon
来源：
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 2000年 / 29卷 / 01期
关键词：
autoregressive processes; least squares estimate; generalized M-estimate; median substitute estimate; additive outliers;
D O I：
10.1080/03610920008832467
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
Estimation of the parameter phi of an autoregressive process of order one, {X-t}, is considered. An estimation procedure referred to as the Median Substitute (MS) method based on the empirical median of {Xt-1Xt} is proposed. phi(MS), the MS estimate of phi, is proven to be consistent and asymtotically normal. A simulation study is carried out to show the viability of the proposed method in comparison with some other robust procedures.
引用
收藏
页码：45 / 54
页数：10
相关论文
共 50 条
12345