13
Bayesian Approach to Nonlinear Mixed-Effects Quantile Regression Models for Longitudinal Data with Non-normality and Left-censoring
Authors: Yangxin Huang, Jiaqing Chen, Xiaosun Lu
Number of views: 606
In longitudinal studies, measurements of the same individuals are taken repeatedly
through time, but it often happens that some collected data are observed with the following issues.
(i) Often, the primary goal is to characterize the change in response over time. Compared with
conventional mean regression, quantile regression (QR) can characterize the entire conditional
distribution of the outcome variable, and may be more robust to outliers and mis-specification
of error distribution. (ii) longitudinal outcomes may suffer from a serious departure of normality
in which normality assumption may cause lack of robustness and subsequently lead to invalid
inference; and (iii) the response observations may be subject to left-censoring due to a limit of
detection. Inferential procedures will become very complicated when one analyzes data with these
features together. In this article, Bayesian modeling approach to nonlinear mixed-effects quantile
regression models for longitudinal data is developed to study simultaneous impact of multiple data
features (non-normality, left-censoring, non-linearity, outliers and heavy-tails). Simulation studies
are conducted to assess the performance of the proposed models and methods. A real data example
is analyzed to demonstrate the proposed methodology through comparing potential models with
different distribution specifications of random-effects.