ScorecardProfet: Scorecard Builder

In Rprofet: WOE Transformation and Scorecard Builder

Scorecard Builder

Description

Function that fits a logistic regression models and scores points for each bin and calculates observations' total score.

Usage

ScorecardProfet(
  object,
  id,
  target,
  GLModel,
  PDO = 100,
  BaseOdds = 10,
  BasePts = 1000,
  reverse = FALSE
)

Arguments

object	A WOEProfet object or a Var_select object that containing dataframes with binned and WOE values.
id	ID variable.
target	A binary target variable.
GLModel	A generalized linear model, glm object.
PDO	Points to Double Odds.
BaseOdds	Base Odds.
BasePts	Base Points.
reverse	Logical. If FALSE, higher points corresponds to a lower probability of being target.

Value

A scorecard dataframe.

Examples

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

binned <- BinProfet(mydata, id= "ID", target= "default", num.bins = 5) ## Binning variables

WOE_dat <- WOEProfet(binned, "ID","default", 3:5) ## WOE transformation of bins

md <- glm(default ~ student_WOE+balance_WOE+income_WOE, data=WOE_dat$WOE, family="binomial")
summary(md)

Score_dat <- ScorecardProfet(object=WOE_dat, id="ID", target="default", GLModel=md,
                                 PDO = 50, BaseOdds = 10, BasePts = 1000, reverse = FALSE)

Score_dat ## Less points means more likely to default

Rprofet documentation built on May 29, 2024, 3:15 a.m.