CollapseLevels: Collapses Levels, Computes Information Value and WoE

An update to the previous version ( 0.1.0 ) of the package. Functions in the previous version required the binary outcome variable of the data set to be strictly a factor.In version 0.2.0 the binary outcome variable may be both a factor (with two levels) or an integer (or numeric ). All functions of the previous version have been retained.No new functions have been added. Like the previous version,provides functions to collapse the levels of an attribute based on response rates. It also provides functions to compute and display information value, and weight of evidence (WoE) for the attributes, and to convert numeric variables to categorical ones by binning. These functions only work for binary classification problems.

Version: 0.2.0
Depends: magrittr
Imports: dplyr, lazyeval, ggplot2
Suggests: knitr, rmarkdown
Published: 2017-12-04
Author: Krishanu Mukherjee
Maintainer: Krishanu Mukherjee <toton1181 at gmail.com>
License: GPL-2
NeedsCompilation: no
CRAN checks: CollapseLevels results

Downloads:

Reference manual: CollapseLevels.pdf
Vignettes: "CollapseLevels"
Package source: CollapseLevels_0.2.0.tar.gz
Windows binaries: r-devel: CollapseLevels_0.2.0.zip, r-release: CollapseLevels_0.2.0.zip, r-oldrel: CollapseLevels_0.2.0.zip
OS X El Capitan binaries: r-release: CollapseLevels_0.2.0.tgz
OS X Mavericks binaries: r-oldrel: CollapseLevels_0.2.0.tgz
Old sources: CollapseLevels archive

Linking:

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