Bayesian Variable Selection for Latent Class Models
Article first published online: 29 OCT 2010
© 2010, The International Biometric Society
Volume 67, Issue 3, pages 917–925, September 2011
How to Cite
Ghosh, J., Herring, A. H. and Siega-Riz, A. M. (2011), Bayesian Variable Selection for Latent Class Models. Biometrics, 67: 917–925. doi: 10.1111/j.1541-0420.2010.01502.x
- Issue published online: 14 SEP 2011
- Article first published online: 29 OCT 2010
- Received December 2009. Revised July 2010. Accepted August 2010.
- Bayesian model averaging;
- Finite mixture model;
- Markov chain Monte Carlo;
- Multinomial logit model;
- Variable selection
Summary In this article, we develop a latent class model with class probabilities that depend on subject-specific covariates. One of our major goals is to identify important predictors of latent classes. We consider methodology that allows estimation of latent classes while allowing for variable selection uncertainty. We propose a Bayesian variable selection approach and implement a stochastic search Gibbs sampler for posterior computation to obtain model-averaged estimates of quantities of interest such as marginal inclusion probabilities of predictors. Our methods are illustrated through simulation studies and application to data on weight gain during pregnancy, where it is of interest to identify important predictors of latent weight gain classes.