Online Program Home
My Program

Abstract Details

Activity Number: 439
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #319933
Title: Multilevel Bayesian Latent Class Growth Mixture Model for Longitudinal Zero-Inflated Poisson Data
Author(s): Kejia Wang* and Kiros Berhane
Companies: University of Southern California and University of Southern California
Keywords: Multilevel ; Latent Class Growth Mixture model ; Zero-inflated Poisson ; Bayesian

There has been a growing interest among researchers in the use of Latent Class Growth mixture modeling techniques for applications in the field of developmental psychology, social and behavioral science. Latent Class Growth Mixture (LCGM) model has been increasingly recognized for its effectiveness to identify distinct subgroups of trajectories based on longitudinal data. Current LCGM methods (e.g., using Mplus and SAS Proc Traj) for Zero-inflated count data, have problems with convergence issue, non-identification, local maxima etc.. This paper focuses on developing multilevel LCGM for Zero-inflated Poisson data. The goal is to model longitudinal growth curves on counts of rare events and to classify individuals into distinct latent groups that represent the unique phenotype of each individual. Bayesian Metropolis algorithm is used to estimate the latent variables and regression parameters in this proposed model. The methodological details are illustrated through a simulation study and are applied to both individual-level data and community-level data on cigarette smoking from the National Longitudinal Study of Adolescent to Adult Health.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association