|
Activity Number:
|
508
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #302391 |
|
Title:
|
A Parametric Mixture Model Method for Clustering Multivariate Correlated Binary Data
|
|
Author(s):
|
Ajit C. Tamhane*+ and Dingxi Qiu
|
|
Companies:
|
Northwestern University and University of Miami
|
|
Address:
|
Department of IEMS, Evanston, IL, 60208-3119,
|
|
Keywords:
|
Latent class analysis ; Clustering ; Multivariate Bernoulli Data ; Maximum likelihood ; Bayes classification rule ; Mixture model
|
|
Abstract:
|
The traditional latent class analysis (LCA) uses a binary mixture model with independent responses. In many practical applications, the responses are locally dependent because they are observed on the same subject. We extend the LCA model to allow for local dependence by fitting a parametric mixture model in which each cluster follows a multivariate Bernoulli distribution. An extension of a family of parametric models by Oman and Zucker (2001) is adopted for this purpose and the method of maximum likelihood estimation is used for fitting. The Bayesian information criterion (BIC) due to Schwarz (1978) is employed to select the number of clusters. Subjects are classified to clusters using the Bayes rule. The proposed mixture model method is illustrated by applying it to two real data sets.
|
