Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing Data

• Published in: Biometrics Vol. 66; no. 1; pp. 105 - 114
• Main Authors: Yuan, Ying, Yin, Guosheng
• Format: Journal Article
• Language: English
• Published: Malden, USA Blackwell Publishing Inc 01.03.2010; Wiley-Blackwell; Blackwell Publishing Ltd
• Subjects: Algorithms; Bayes Theorem; Bayesian analysis; Bayesian inference; BIOMETRIC METHODOLOGY; Biometrics; biometry; Biometry - methods; clinical trials; Computer Simulation; Data Interpretation, Statistical; Data models; Epidemiologic Methods; Error analysis; Estimation bias; Information Storage and Retrieval - methods; Informative missing data; Longitudinal data; Longitudinal Studies; Measurement techniques; Missing data; Modeling; Models, Statistical; Multilevel models; Nonignorable dropout; Penalized function; Quantile regression; Random effects; Regression analysis; Repeated measures; Sample Size; School dropouts; Shared-parameter model; Statistical analysis
• ISSN: 0006-341X; 1541-0420; 1541-0420
• DOI: 10.1111/j.1541-0420.2009.01269.x

We study quantile regression (QR) for longitudinal measurements with nonignorable intermittent missing data and dropout. Compared to conventional mean regression, quantile regression can characterize the entire conditional distribution of the outcome variable, and is more robust to outliers and misspecification of the error distribution. We account for the within-subject correlation by introducing a [graphic removed] penalty in the usual QR check function to shrink the subject-specific intercepts and slopes toward the common population values. The informative missing data are assumed to be related to the longitudinal outcome process through the shared latent random effects. We assess the performance of the proposed method using simulation studies, and illustrate it with data from a pediatric AIDS clinical trial.