WOE_customNum {Rprofet}

Custom Binning Numeric Variables

Description

Function that bins a numeric variable based on user inputted breaks, plots the information on the new bins, and returns a list contains a dataframe of the newly binned values and id column and more items.

Usage

WOE_customNum(
  data,
  var,
  id,
  target,
  breaks,
  right_bracket = F,
  color = "#0066CC",
  plot = FALSE
)

Arguments

data

Dataframe containing the target variable and desired numeric variables to be binned.

var

A specific numeric attribute to be binned. Must be specified.

id

The unique id variable in the dataframe. Must be specified.

target

A binary target variable. Must be specified.

breaks

A vector of breakpoints for the desired bins. Must be specified.

right_bracket

Logical. Specifying whether the intervals are closed on the right or the left.

color

A hexadecimal value representing a specific color.

plot

Logical. The default is FALSE which does not generate the plots.

Value

A list with the following components.

NewBin

Dataframe with the binned variable.

BinWOE

Dataframe with target, binned variable, and WOE values for the bins.

IV

Information value of the newly binned variable.

vars

Dataframe with binned variable, WOE values for the bins, Target Rate for each bin, and observation count for each bin.

Examples

mydata <- ISLR::Default
mydata$ID = seq(1:nrow(mydata)) ## make the ID variable
mydata$default <- ifelse(mydata$default=="Yes", 1, 0) ## target coded with 1, 0

WC_1 <- WOE_customNum(data= mydata, var="balance", id= "ID", target = "default",
                      breaks= seq(0,3000,1000))
head(WC_1$NewBin)
head(WC_1$BinWOE)
WC_1$IV
WC_1$vars

WC_2 <- WOE_customNum(data= mydata, var="income", id= "ID", target = "default",
                      breaks=seq(0,75000, 15000))
head(WC_2$NewBin)
head(WC_2$BinWOE)
WC_2$IV
WC_2$vars