APPROXIMATE METHODS USING RANKS FOR REGRESSION WITH CENSORED-DATA
成果类型：
Article
署名作者：
PETTITT, AN
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.2307/2335949
发表日期：
1983
页码：
121132
关键词：
摘要：
An approximation to marginal rank likelihoods is given and used to make inferences for the linear regression model with censored data. The approximations involve linear rank statistics, adapted for censored data, and estimates of their variance. Various explicit scores are given for right censored data and logistic scores for doubly censored data. Inferences using the approximate rank analysis are compared with inferences using fully and partially parametric techniques on 2 data sets. [An example is given using heart transplant data.].