Second-order biases of maximum likelihood estimates in overdispersed generalized linear models

被引：7
|
作者：
Cordeiro, GM
Botter, DA
机构：
[1] UFBA, Dept Estat, BR-40170110 Salvador, BA, Brazil
[2] USP, Dept Estat, BR-05315970 Sao Paulo, SP, Brazil
来源：
STATISTICS & PROBABILITY LETTERS | 2001年 / 55卷 / 03期
基金：
巴西圣保罗研究基金会;
关键词：
bias correction; exponential family; generalized linear model; link function; maximum likelihood estimate; overdispersion;
D O I：
10.1016/S0167-7152(01)00150-X
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 derive general formulae for second-order biases of maximum likelihood estimates in overdispersed generalized linear models, thus generalizing results by Cordeiro and McCullagh (J. Roy. Statist. Soc. Ser. B 53 (1991) 629), and Botter and Cordeiro (Statist. Comput. Simul. 62 (1998) 91). Our formulae cover many important and commonly used models and are easily implemented by means of supplementary weighted linear regressions. They are also simple enough to be used algebraically to obtain several closed-form expressions in special models. The practical use of such formulae is illustrated in a simulation study. (C) 2001 Elsevier Science B.V. All rights reserved.
引用
收藏
页码：269 / 280
页数：12
相关论文
共 50 条
12345