globalboosttest: Testing the additional predictive value of high-dimensional data

'globalboosttest' implements a permutation-based testing procedure to globally test the (additional) predictive value of a large set of predictors given that a small set of predictors is already available. Currently, 'globalboosttest' supports binary outcomes (via logistic regression) and survival outcomes (via Cox regression). It is based on boosting regression as implemented in the package 'mboost'.

Version: 1.1-0
Depends: R (≥ 2.8), mboost (≥ 2.0-0), survival
Published: 2012-10-29
Author: Anne-Laure Boulesteix, Torsten Hothorn.
Maintainer: Anne-Laure Boulesteix <boulesteix at ibe.med.uni-muenchen.de>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
In views: Survival
CRAN checks: globalboosttest results

Downloads:

Reference manual: globalboosttest.pdf
Package source: globalboosttest_1.1-0.tar.gz
Windows binaries: r-devel: globalboosttest_1.1-0.zip, r-release: globalboosttest_1.1-0.zip, r-oldrel: globalboosttest_1.1-0.zip
OS X Snow Leopard binaries: r-release: globalboosttest_1.1-0.tgz, r-oldrel: globalboosttest_1.1-0.tgz
OS X Mavericks binaries: r-release: globalboosttest_1.1-0.tgz