Distribution-Free Models for Latent Population Mixtures

• Activity Number: 301
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Hui Zhang*+
• Companies: University of Rochester Medical Center
• Keywords: Zero-Inflated Poisson ; mixture ; distribution-free ; functional response model ; latent ; inverse probability weighted

Abstract

Many studies involve mixtures of latent populations. With the unobservable sub-groups, it is not possible to compare intervention effects across them using existing methods. In this talk, we propose a novel approach to tackle the analytic problems. We will employ a Zero-inflated Poisson like model to help identify the two latent sub-groups and integrate this sub-model into the context of a primary model of interest to enable estimation of parameters of interest. We develop this proposed system of models by utilizing a new class of functional response models(FRM). To provide inference for longitudinal data analysis, we integrate the inverse probability weighted estimate within the context of FRM and develop distribution-free inference about the parameters of the system. The proposed distribution-free approach seems to offer the only sensible solution to this type of modeling problem.