Online Program Home
My Program

Abstract Details

Activity Number: 156
Type: Topic Contributed
Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
Sponsor: Section on Nonparametric Statistics
Abstract #319723 View Presentation
Title: Nonparametric Identifiability of Finite Mixture Model and It's Application in Alzheimer's Disease
Author(s): Zheyu Wang* and Xiao-Hua Andrew Zhou
Companies: The Johns Hopkins University and University of Washington
Keywords: Finite mixture model ; Identifiability ; Latent variables

Finite mixture model have long been utilized to summarize constructs that are represented by multiple variables and are difficult to measure. It also had significant application and development in biomarker evaluation where ascertaining the true disease status can be too costly, too invasive, or too late for any treatment intervention to be effective and therefore unavailable at the time of assessment. This talk will discuss how to adopt a finite mixture model in such situation that allows for flexible component distribution as well as covariate effects that may affect component distribution or the mixing proportion. The talk will also introduce a permutation identifiability concept to discuss the nonparametric global identifiability for this type of model in general.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association