LEVERAGE AND COCHRANE-ORCUTT ESTIMATION IN LINEAR-REGRESSION

被引：10

作者：

STEMANN, D ^{[1
]}

TRENKLER, G ^{[1
]}

机构：

[1] UNIV DORTMUND,DEPT STAT,W-4600 DORTMUND 50,GERMANY

来源：

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 1993年 / 22卷 / 05期

关键词：

REGRESSION; COCHRANE-ORCUTT ESTIMATOR; AUTOCORRELATION; LEVERAGE; EFFICIENCY;

D O I：

10.1080/03610929308831088

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：

In this paper we investigate the influence of the correlation coefficient on the hat matrix diagonal component corresponding to the first transformed observation in regression models with AR(1)-errors. Furthermore a relationship between this first leverage and the efficiency of the Cochrane-Orcutt Estimator is established.

引用

页码：1315 / 1333

页数：19

共 9 条

[1]

ANDREWS DF, 1978, J ROY STAT SOC B MET, V40, P85

[2]

Belsley D. A., 2005, REGRESSION DIAGNOSTI

[3]

BERENBLUT II, 1973, J R STAT SOC B, V35, P33

[4]

BHARGAVA A, 1989, BIOMETRIKA, V76, P823