The regression-calibration method for fitting generalized linear models with additive measurement error

被引：39

作者：

Hardin, James W. ^{[1
]}

Schmiediche, Henrik ^{[2
]}

Carroll, Raymond J. ^{[2
]}

机构：

[1] Univ South Carolina, Arnold Sch Publ Hlth, Columbia, SC 29208 USA

[2] Texas A&M Univ, Dept Stat, College Stn, TX 77843 USA

来源：

STATA JOURNAL | 2003年 / 3卷 / 04期

基金：

美国国家卫生研究院;

关键词：

st0050; regression calibration; measurement error; instrumental variables; replicate measures; generalized linear models;

D O I：

10.1177/1536867X0300300406

中图分类号：

O1 [数学]; C [社会科学总论];

学科分类号：

03 ; 0303 ; 0701 ; 070101 ;

摘要：

This paper discusses and illustrates the method of regression calibration. This is a straightforward technique for fitting models with additive measurement error. We present this discussion in terms of generalized linear models (GLMs) following the notation defined in Hardin and Carroll (2003). Discussion will include specified measurement error, measurement error estimated by replicate error-prone proxies, and measurement error estimated by instrumental variables. The discussion focuses on software developed as part of a small business innovation research (SBIR) grant from the National Institutes of Health (NIH).

引用

页码：361 / 372

页数：12

共 7 条

[1] MEASUREMENT ERROR IN THE GENERALIZED LINEAR-MODEL [J].

ARMSTRONG, B .

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 1985, 14 (03) :529-544

[2] APPROXIMATE QUASI-LIKELIHOOD ESTIMATION IN MODELS WITH SURROGATE PREDICTORS [J].

CARROLL, RJ ;

STEFANSKI, LA .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1990, 85 (411) :652-663

[3]

Carroll RJ., 1995, MEASUREMENT ERROR NO

[4]

Glesjer L. J., 1990, STAT ANAL MEASUREMEN

[5]

Hardin J., 2003, STATA J, V3, P372

[6]

Hardin J., 2003, STATA J, V3, P328

[7] COVARIATE MEASUREMENT ERRORS AND PARAMETER-ESTIMATION IN A FAILURE TIME REGRESSION-MODEL [J].

PRENTICE, RL .

BIOMETRIKA, 1982, 69 (02) :331-342

← 1 →