Quantile Regression Methods for Doubly Censored Data from Registry Studies

• Activity Number: 624
• Type: Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Shuang Ji*+ and Limin Peng+ and Yu Cheng and Huichuan Lai
• Companies: Emory University and Emory University and University of Pittsburgh and University of Wisconsin
• Keywords: quantile regression ; survival analysis ; doubly censored data ; martingale ; varying-effects model

Abstract

Quantile regression is known for its flexibility of accommodating varying covariate effects and has attracted growing interests in its application to survival analysis. Based on Peng and Huang (2008)'s work on quantile regression method with randomly right censored data, we develop a quantile regression method for a double censoring setting often encountered in registry studies utilizing the embedded martingale structure. The proposed estimation and inference procedures are computationally simple and stable. We establish the uniform consistency and weak convergence of the resulting estimators. We also provide a sensible solution to address the identifiability issues on regression quantiles at both tails, a unique feature with doubly censored data. The finite-sample performance of our approach is assessed by a series of simulation studies. An application to a registry data on cystic fibr