Resampling procedures to assess the stability of selected variables with additional finite sample error control for high-dimensional variable selection procedures such as Lasso or boosting. Both, standard stability selection (Meinshausen & Buhlmann, 2010, ) and complementary pairs stability selection with improved error bounds (Shah & Samworth, 2013, ) are implemented. The package can be combined with arbitrary user specified variable selection approaches.

Maintainer: Benjamin Hofner <benjamin.hofner at pei.de>

Author(s): Benjamin Hofner*, Torsten Hothorn*

Install package and any missing dependencies by running this line in your R console:

install.packages("stabs")

Depends R (>= 2.14.0), methods, stats, parallel
Imports graphics, grDevices, utils
Suggests glmnet, lars, mboost(>2.3-0), gamboostLSS(>=1.2-0), QUIC, TH.data, hdi, testthat, knitr, rmarkdown, igraph, huge
Enhances
Linking to
Reverse
depends
gamboostLSS, mboost
Reverse
imports
DIFboost, FDboost
Reverse
suggests
Reverse
enhances
Reverse
linking to

Package stabs
Materials
URL https://github.com/hofnerb/stabs
Task Views MachineLearning
Version 0.6-2
Published 2017-01-31
License GPL-2
BugReports
SystemRequirements
NeedsCompilation no
Citation
CRAN checks stabs check results
Package source stabs_0.6-2.tar.gz