rfVarImpOOB: Unbiased Variable Importance for Random Forests

Computes a novel variable importance for random forests: Impurity reduction importance scores for out-of-bag (OOB) data complementing the existing inbag Gini importance, see also <doi:10.1080/03610926.2020.1764042>. The Gini impurities for inbag and OOB data are combined in three different ways, after which the information gain is computed at each split. This gain is aggregated for each split variable in a tree and averaged across trees.

Version: 1.0.1
Depends: R (≥ 3.2.2), stats, randomForest
Imports: ggplot2, binaryLogic, dplyr, titanic, prob, ggpubr, magrittr
Suggests: knitr, rmarkdown
Published: 2020-10-18
Author: Markus Loecher
Maintainer: Markus Loecher <Markus.Loecher at gmail.com>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Citation: rfVarImpOOB citation info
CRAN checks: rfVarImpOOB results


Reference manual: rfVarImpOOB.pdf
Vignettes: Unbiased Variable Importance
Package source: rfVarImpOOB_1.0.1.tar.gz
Windows binaries: r-devel: rfVarImpOOB_1.0.zip, r-release: rfVarImpOOB_1.0.1.zip, r-oldrel: rfVarImpOOB_1.0.1.zip
macOS binaries: r-release: rfVarImpOOB_1.0.1.tgz, r-oldrel: rfVarImpOOB_1.0.1.tgz
Old sources: rfVarImpOOB archive


Please use the canonical form https://CRAN.R-project.org/package=rfVarImpOOB to link to this page.