Online Program Home
My Program

Abstract Details

Activity Number: 311
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract #321036 View Presentation
Title: A Finite Mixture Model for Clustered Bivariate Binary Data: Application to Ophthalmologic Data Structures
Author(s): John Kwagyan*
Companies: Howard University College of Medicine
Keywords: Bivariate binary data ; Clustered data ; Finite Mixture Models ; mixed-effects models ; Ophthalmologic Data ; Zero-inflated models

Ophthalmologic data have traditionally posed challenges for statistical modelling and inference. First there is the potential association between pairs of eyes. Then there is the situation where data is available on one eye for some persons and on both eyes for others. Measurement of multivariate outcomes also occur time and again in ophthalmologic studies, usually because the diseases are related or form a constellation as a syndrome. It is often of clinical interest to model the inter-correlation not only between an outcome and risk factors, but also between different outcomes. We develop a computationallytractable likelihood-based approach that would allow for the detection of correlation between bivariate dichotomous outcomes, modeled simultaneously with the between-eye correlation with and without covariate effects.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association