Quantile regression for doubly truncated data

Abstract

Doubly truncated data arise when the event time of interest T is observed only if it falls within a subject-specific, possibly random, interval . In this article, we study the problem of fitting a quantile regression model with doubly truncated data. Based on the non-parametric maximum likelihood estimator of , we proposed a weighted quantile regression estimator. Our method leads to a simple algorithm that can be conveniently implemented with R software. We show that the proposed estimator is consistent and asymptotically normal under appropriate conditions. We evaluate the finite sample performance of the proposed estimators through simulation studies. The proposed method is illustrated using AIDS blood transfusion data.