SIMEX method for censored quantile regression with measurement error

ABSTRACT

Censored quantile regression serves as an important supplement to the Cox proportional hazards model in survival analysis. In addition to being exposed to censoring, some covariates may subject to measurement error. This leads to substantially biased estimate without taking this error into account. The SIMulation-EXtrapolation (SIMEX) method is an effective tool to handle the measurement error issue. We extend the SIMEX approach to the censored quantile regression with covariate measurement error. The algorithm is assessed via extensive simulations. A lung cancer study is analyzed to verify the validation of the proposed method.

KEYWORDS:

• Measurement error
• Quantile regression
• SIMEX method
• Survival data

MATHEMATICS SUBJECT CLASSIFICATION:

• 62 (Statistics)
• 62N02 (Estimation)

Acknowledgments

The authors are grateful for the valuable comments and suggestions from the associate editor and the referees, which drastically improved the article.

Funding

This work is supported by the National Science Foundation of China grant NSFC 11571263 and NSFC 11371299.