clustvarsel: Variable Selection for Model-Based Clustering

A function which implements variable selection methodology for model-based clustering which allows to find the (locally) optimal subset of variables in a dataset that have group/cluster information. A greedy or headlong search can be used, either in a forward-backward or backward-forward direction, with or without sub-sampling at the hierarchical clustering stage for starting Mclust models. By default the algorithm uses a sequential search, but parallelization is also available.

Version: 2.0
Depends: R (≥ 3.0.0), mclust (≥ 4.0), BMA (≥ 3.16), foreach, iterators
Suggests: MASS, parallel, doParallel
Published: 2013-10-25
Author: Nema Dean, Adrian E. Raftery, and Luca Scrucca
Maintainer: Luca Scrucca <luca at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Citation: clustvarsel citation info
Materials: NEWS
In views: ChemPhys, Cluster, Multivariate
CRAN checks: clustvarsel results


Reference manual: clustvarsel.pdf
Vignettes: clustvarsel: A Package Implementing Variable Selection for Model-based Clustering
Package source: clustvarsel_2.0.tar.gz
MacOS X binary: clustvarsel_2.0.tgz
Windows binary:
Old sources: clustvarsel archive