Testing for varying dispersion in exponential family nonlinear models

被引：34

作者：

Wei, BC

Shi, JQ

Fung, WK

Hu, YQ

机构：

[1] Southeast Univ, Dept Math, Nanjing 210096, Peoples R China

[2] Chinese Univ Hong Kong, Dept Stat, Hong Kong, Hong Kong

[3] Univ Hong Kong, Dept Stat, Hong Kong, Peoples R China

来源：

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS | 1998年 / 50卷 / 02期

关键词：

adjusted profile likelihood; exponential family nonlinear models; gamma model; inverse Gaussian model; score statistic; simulation study; varying dispersion;

D O I：

10.1023/A:1003491131768

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：

A diagnostic model and several new diagnostic statistics are proposed for testing for varying dispersion in exponential family nonlinear models. A score statistic and an adjusted score statistic based on Cox and Reid (1987, J. Roy. Statist. Sec. Ser. B, 55, 467-471) are derived in normal, inverse Gaussian! and gamma nonlinear models. An adjusted likelihood ratio statistic is also given for normal and inverse Gaussian nonlinear models. The results of simulation-studies are presented, which show that the adjusted tests keep their sizes better and are more powerful than the ordinary tests.

引用

页码：277 / 294

页数：18

共 24 条

[1]

[Anonymous], 1993, The Inverse Gaussian Distribution: A Case Study in Exponential Families"

[2] TESTS OF HYPOTHESES IN OVERDISPERSED POISSON REGRESSION AND OTHER QUASI-LIKELIHOOD MODELS [J].