-
Notifications
You must be signed in to change notification settings - Fork 0
/
DESCRIPTION
25 lines (25 loc) · 996 Bytes
/
DESCRIPTION
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Package: miv.select
Title: A package for aided manual variable selection in binary logistic regression based models using marginal information value of features.
Version: 0.0.1
Authors@R: person("Louis", "Vines", email = "[email protected]", role = c("aut", "cre"))
Description:
This package builds methods to aid with binary classification through logistic regression based methods (including GLMNet and GAMS), particularly focused on applications in credit scoring. It provides methods for supervised and unsupervised binning, single factor screening using information values and a variable selection based on marginal information values.
Depends: R (>= 3.3.3)
Imports:
dplyr (>= 0.7.2),
ggplot2 (>= 2.2.1),
partykit (>= 1.0.5),
purrr (>= 0.2.2),
stringr (>= 1.2.0),
forcats (>= 0.2.0),
cowplot (>= 0.8.0),
ROCR (>= 1.0-7)
Suggests:
testthat,
knitr,
rmarkdown
License: GPL (>= 2)
Encoding: UTF-8
LazyData: true
RoxygenNote: 6.0.1
VignetteBuilder: knitr