CorReg: Linear regression based on linear structure between covariates

Linear regression based on a recursive structural equation model (explicit correlations) found by a MCMC algorithm. It permits to face highly correlated covariates. Variable selection is included (by lasso, elasticnet, etc.). It also provides some graphical tools for basic statistics.

Version: 0.15.8
Depends: R (≥ 3.0.2)
Imports: Rcpp (≥ 0.11.0), lars (≥ 1.2), elasticnet (≥ 1.1), corrplot (≥ 0.73), Matrix (≥ 1.1), ridge (≥ 2.1), rpart (≥ 4.1-5), MASS (≥ 7.3-30), mvtnorm (≥ 0.9)
LinkingTo: Rcpp, RcppEigen
Suggests: clere (≥ 1.1), spikeslab (≥ 1.1.5), parcor (≥ 0.2), Rmixmod (≥ 2.0.1), rtkpp (≥ 0.8.3), mclust (≥ 4.2)
Published: 2014-11-04
Author: Clement THERY [aut, cre], Christophe BIERNACKI [ctb], Gaetan LORIDANT [ctb], Florian WATRIN [ctb], Quentin GRIMONPREZ [ctb], Vincent KUBICKI [ctb], Samuel BLANCK [ctb], Jeremie KELLNER [ctb]
Maintainer: Clement THERY <clement.thery at arcelormittal.com>
License: CeCILL
Copyright: ArcelorMittal
URL: http://www.correg.org
NeedsCompilation: yes
Materials: NEWS
CRAN checks: CorReg results

Downloads:

Reference manual: CorReg.pdf
Package source: CorReg_0.15.8.tar.gz
Windows binaries: r-devel: CorReg_0.15.8.zip, r-release: CorReg_0.15.8.zip, r-oldrel: CorReg_0.15.8.zip
OS X Snow Leopard binaries: r-release: CorReg_0.15.8.tgz, r-oldrel: CorReg_0.15.8.tgz
OS X Mavericks binaries: r-release: CorReg_0.15.8.tgz
Old sources: CorReg archive