Premium
Estimation and hypothesis testing with error‐contaminated survival data under possibly misspecified measurement error models
Author(s) -
Yi Grace Y.,
Yan Ying
Publication year - 2021
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.1002/cjs.11594
Subject(s) - covariate , observational error , statistics , wald test , estimation , econometrics , statistical hypothesis testing , errors in variables models , computer science , process (computing) , mathematics , engineering , systems engineering , operating system
In the presence of covariate measurement error, there has been extensive interest in developing estimation methods for parameters associated with various survival models, where the classical additive measurement error model is commonly used to describe the measurement error process. On the contrary, hypothesis testing has been less explored for survival data with error‐contaminated covariates. Furthermore, it is important to study the impact of misspecification of the measurement error process. In this article, we propose a “corrected” score test and a “corrected” Wald test and establish their theoretical properties. Moreover, we exploit the impact of misspecification of measurement error models on parameter estimation and hypothesis testing. Simulation studies are reported to demonstrate the finite‐sample performance of the proposed methods, and a real data example is presented to illustrate the usage of our methods.