|
Title
|
|
Introduction to Latent Class Mixture Models
|
|
Date / Time / Room
|
Sponsor
|
Type
|
08/13/2002
8:00 AM -
12:00 PM
Room: H-Lincoln Suite
|
ASA, Section on Statistical Graphics*
|
Other
|
|
Organizer:
|
n/a
|
|
Chair:
|
n/a
|
|
CE Presenter
|
Jeroen K. Vermunt - Tilburg University
Jeroen K. Vermunt - Tilburg University
- Statistical Innovations, Inc.
- Statistical Innovations, Inc.
|
Description
Interest in and use of latent class (LC) and finite mixture models is growing rapidly in the social and biomedical sciences because of 1) major developments in maximum likelihood computational algorithms for these models and 2) the lack of restrictive assumptions underlying the general latent class model. In this half-day course we introduce LC as a probability model and focus on 3 important special cases -- LC segmentation/cluster analysis, LC factor analysis, and LC regression analysis, for combinations of nominal, ordinal, and continuous variables. Graphical displays of results will be emphasized.
Both traditional and graphical kinds of output are implemented in a new computer program called Latent GOLD which will be used for demonstrations. The only prerequisite is familiarity with traditional applications of cluster, factor and regression analysis and the chi-squared test for testing the fit of a model. A demo version of the program, sample tutorials and reprints of technical publications can be obtained from the website www.latentclass.com (Copies of the 200 page Latent GOLD manual are available for purchase for use as an optional text for this workshop.).
|
|
