Adjusted regression estimation for time-to-event data with differential measurement error. (English) Zbl 1284.62441

Yu, Menggang

Biometrika 100, No. 3, 757-763 (2013).

Summary: Differential measurement error data plausibly arise in epidemiology and biomedical studies but have been rarely dealt with explicitly, especially for time-to-event data. We propose an estimation equation correction method in semiparametric censored linear regression to deal with differential measurement error for time-to-event data with validation samples. The method does not require explicit modelling of the error-prone covariates and leads to unbiased estimation.

MSC:

62J05	Linear regression; mixed models
62G08	Nonparametric regression and quantile regression
62G20	Asymptotic properties of nonparametric inference
65C60	Computational problems in statistics (MSC2010)

Keywords:

censored linear regression; model misspecification; validation sample