CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Multivariate-t nonlinear mixed models for censored multi-outcome longitudinal data Authors:  Wan-Lun Wang - Feng Chia University (Taiwan) [presenting]
Tsung-I Lin - National Chung Hsing University (Taiwan)
Abstract: In multivariate longitudinal studies, multi-outcome repeated measures on each subject over time may contain outliers, and the responses are often subject to an upper or lower limit of detection depending on the quantification assays. We consider an extension of the multivariate nonlinear mixed effects model by adopting a joint multivariate-$t$ distribution for random effects and within-subject errors and taking the censoring information of multiple responses into account. The proposed model, called the multivariate-$t$ nonlinear mixed model with censored responses (MtNLMMC), allows for analyzing multi-outcome longitudinal data exhibiting nonlinear growth patterns with censorship and fat-tailed behavior. Utilizing the Taylor-series linearization method, a pseudo-data version of expectation conditional maximization either (ECME) algorithm is developed for iteratively carrying out maximum likelihood estimation. We demonstrate our methods with HIV/AIDS data examples and simulation studies. Experimental results signify that the MtNLMMC performs favorably compared to its normal analogue and some existing approaches.