Least absolute deviation estimation for regression with ARMA errors

被引：63

作者：

Davis, RA ^{[1
]}

Dunsmuir, WTM ^{[1
]}

机构：

[1] UNIV NEW S WALES,KENSINGTON,NSW 2033,AUSTRALIA

来源：

JOURNAL OF THEORETICAL PROBABILITY | 1997年 / 10卷 / 02期

关键词：

ARMA process; regression; least absolute deviation estimation; central limit theorem;

D O I：

10.1023/A:1022620818679

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：

The asymptotic normality for least absolute deviation estimates of the parameters in a linear regression model with autoregressive moving average errors is established under very general conditions. The method of proof is based on a functional limit theorem for the LAD objective function.

引用

页码：481 / 497

页数：17

共 10 条

[1]

Bloomfield P., 1983, LEAST ABSOLUTE DEVIA

[2]

BROCKWELL PJ, 1991, TIME SERIESA THEORY

[3] Maximum likelihood estimation for MA(1) processes with a root on or near the unit circle [J].