SelvarMix: Regularization for Variable Selection in Model-Based Clustering and Discriminant Analysis

Performs a regularization approach to variable selection in the model-based clustering and classification frameworks. First, the variables are arranged in order with a lasso-like procedure. Second, the method of Maugis, Celeux, and Martin-Magniette (2009, 2011) is adapted to define the role of variables in the two frameworks.

Version: 1.1
Imports: Rcpp (≥ 0.11.1), glasso, parallel, Rmixmod, methods
LinkingTo: Rcpp, RcppArmadillo
Published: 2015-09-20
Author: Mohammed Sedki, Gilles Celeux, Cathy Maugis-Rabusseau
Maintainer: Mohammed Sedki <mohammed.sedki at u-psud.fr>
License: GPL (≥ 3)
NeedsCompilation: yes
Materials: README
CRAN checks: SelvarMix results

Downloads:

Reference manual: SelvarMix.pdf
Package source: SelvarMix_1.1.tar.gz
Windows binaries: r-devel: SelvarMix_1.1.zip, r-release: SelvarMix_1.1.zip, r-oldrel: SelvarMix_1.1.zip
OS X Snow Leopard binaries: r-release: SelvarMix_1.1.tgz, r-oldrel: SelvarMix_1.0.tgz
OS X Mavericks binaries: r-release: SelvarMix_1.1.tgz
Old sources: SelvarMix archive