Variable Selection for Latent Class Analysis

Variable selection for latent class analysis for model-based clustering of multivariate categorical data. The package implements a general framework for selecting the subset of variables with relevant clustering information and discard those that are redundant and/or not informative. The variable selection method is based on the approach of Fop et al. (2017) and Dean and Raftery (2010) . Different algorithms are available to perform the selection: stepwise, swap-stepwise and evolutionary stochastic search. Concomitant covariates used to predict the class membership probabilities can also be included in the latent class analysis model. The selection procedure can be run in parallel on multiple cores machines.


Version 1.1

-- Argument "bicDiff" added to function "LCAvarsel" -- Updated documentation and citation -- Minor bug fixes

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.1 by Michael Fop, 4 years ago

Browse source code at

Authors: Michael Fop [aut, cre] , Thomas Brendan Murphy [ctb]

Documentation:   PDF Manual  

Task views: Cluster Analysis & Finite Mixture Models, Psychometric Models and Methods

GPL (>= 2) license

Imports nnet, MASS, foreach, parallel, doParallel, GA, memoise

Depends on poLCA

Suggests knitr, rmarkdown

See at CRAN