Tail Functions and Iterative Weights in Binary Regression
作者:
MurrayA. Jorgensen,
期刊:
The American Statistician
(Taylor Available online 1994)
卷期:
Volume 48,
issue 3
页码: 230-234
ISSN:0003-1305
年代: 1994
DOI:10.1080/00031305.1994.10476062
出版商: Taylor & Francis Group
关键词: Generalized linear model;Hazard function;Iteratively reweighted least squares;Link function;Logistic regression;Probit regression;Tolerance distribution
数据来源: Taylor
摘要:
In interpreting the binary regression models often used in the analysis of dose-response data, it is common to introduce the idea of an underlying continuous tolerance distribution. Different choices of link function lead to different tolerance distributions. A useful way of comparing these alternatives is to compare the hazard functions or tail functions associated with each tolerance distribution. Tail functions can also be applied to give numerically preferable formulas for the iterative weights and the adjusted dependent variable in the fitting of binary regression models by the iteratively reweighted least-squares algorithm.
点击下载:
PDF (628KB)
返 回