Premium
Additive hazards regression of failure time data with covariate measurement errors
Author(s) -
Sun Liuquan,
Zhang Zhigang,
Sun Jianguo
Publication year - 2006
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/j.1467-9574.2006.00340.x
Subject(s) - covariate , estimator , proportional hazards model , statistics , mathematics , inference , nonparametric statistics , regression analysis , regression , econometrics , observational error , nonparametric regression , computer science , artificial intelligence
This paper discusses regression analysis of failure time data under the additive hazards model, with covariates subject to measurement errors. In particular, we consider the situation where there may exist only repeated measurements or observations on the covariates with measurement errors, and for which there does not exist any method for inference. A nonparametric‐correction approach is proposed for inference about regression parameters and the baseline cumulative hazards function. Both asymptotic and finite sample properties of the proposed estimators are established and the approach is applied to an AIDS clinical trial that motivated this study.