|
Activity Number:
|
81
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
ENAR
|
|
Abstract #318779
|
|
|
Title:
|
An Entropy-Based Model Selection Criterion for Latent Class Analysis of Incomplete Data
|
|
Author(s):
|
Chantal Larose* and Ofer Harel and Katarzyna Kordas and Dipak Dey
|
|
Companies:
|
SUNY New Paltz and University of Connecticut and University of Bristol and University of Connecticut
|
|
Keywords:
|
missing data ;
multiple imputation ;
latent class analysis ;
family studies ;
entropy
|
|
Abstract:
|
Current methods for clustering categorical incomplete data via latent class analysis and multiple imputation tend to only consider the case where the correct number of classes is known prior to imputation. In this talk, a new model selection criterion is presented for LCA of incomplete data where the number of classes is not known prior to imputing missing values. Its performance is compared against AIC and BIC in simulation studies and a family studies data application.
|
Authors who are presenting talks have a * after their name.
