glinternet: Learning interactions via hierarchical group-lasso regularization

Fits linear pairwise-interaction models that satisfy strong hierarchy: if an interaction coefficient is estimated to be nonzero, then its two associated main effects also have nonzero estimated coefficients. Accommodates categorical variables (factors) with arbitrary numbers of levels, continuous variables, and combinations thereof. Implements the machinery described in the paper below.

Version: 0.9.0
Suggests: ggplot2
Published: 2013-08-17
Author: Michael Lim, Trevor Hastie
Maintainer: Michael Lim <michael626 at gmail.com>
License: GPL-2
URL: http://arxiv.org/abs/1308.2719
NeedsCompilation: yes
CRAN checks: glinternet results

Downloads:

Reference manual: glinternet.pdf
Package source: glinternet_0.9.0.tar.gz
OS X binary: glinternet_0.9.0.tgz
Windows binary: glinternet_0.9.0.zip