LCAvarsel: 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) <doi:10.1214/17-AOAS1061> and Dean and Raftery (2010) <doi:10.1007/s10463-009-0258-9>. 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
Depends: R (≥ 3.4), poLCA (≥ 1.4.1)
Imports: nnet, MASS, foreach, parallel, doParallel, GA, memoise
Suggests: knitr (≥ 1.12), rmarkdown (≥ 1.2)
Published: 2018-01-04
Author: Michael Fop [aut, cre], Thomas Brendan Murphy [ctb]
Maintainer: Michael Fop <michael.fop at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Citation: LCAvarsel citation info
Materials: NEWS
CRAN checks: LCAvarsel results


Reference manual: LCAvarsel.pdf
Package source: LCAvarsel_1.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel: not available
OS X El Capitan binaries: r-release: LCAvarsel_1.1.tgz
OS X Mavericks binaries: r-oldrel: not available
Old sources: LCAvarsel archive


Please use the canonical form to link to this page.