FINITE-SAMPLE OPTIMALITY OF MAXIMUM PARTIAL LIKELIHOOD ESTIMATION IN COX MODEL FOR COUNTING-PROCESSES

被引：2

作者：

CHANG, IS

HSIUNG, CA

机构：

[1] NATL CENT UNIV,DEPT MATH,CHUNGLI,TAIWAN

[2] ACAD SINICA,INST STAT SCI,TAIPEI 115,TAIWAN

来源：

JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 1990年 / 25卷 / 01期

关键词：

Estimating equation; Fisher information; martingales; nuisance parameter;

D O I：

10.1016/0378-3758(90)90005-F

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：

Within the framework of estimating function theory, this paper adapts the concept of Fisher information in the presence of a nuisance parameter (Godambe and Thompson (1974), Godambe (1984)) to the context of Cox-type regression model, formulates its Godambe's optimality criterion, and establishes the optimality of the maximum partial likelihood estimation in Cox's model with respect to this criterion. © 1990.

引用

页码：35 / 42

页数：8

共 16 条

[1] COX REGRESSION-MODEL FOR COUNTING-PROCESSES - A LARGE SAMPLE STUDY [J].