liftPlot: Lift Plot

In SixiangHu/DataMan: Data Summary, One-way Interactive Plot and Model Visualisation Analysis

Description

With provided different model predictions, different lift curves will be plot to compare model improvement.

Usage

liftPlot(act,pred1,pred2,weight=NULL,exposure=NULL,breaks=seq(0,2,0.05),showas=NULL)

Arguments

act	A numeric vector to give actual for different.
pred1	A numeric vector to give actual for different.
pred2	A numeric vector to give actual for different.
weight	A numeric vector to give weights for different predictions.
exposure	A numeric vector to give exposure for different predictions.
breaks	A sequence of the numericals defining breaks that differences are split into.
showas	A string vectors to give names for each model on the lift curve.

Author(s)

Sixiang Hu

Examples

glm1 <- glm(mpg~cyl,data=mtcars,family=Gamma(log))
glm2 <- glm(mpg~cyl+vs,data=mtcars,family=Gamma(log))
pred1 <- predict(glm1,mtcars)
pred2 <- predict(glm2,mtcars)
liftPlot(mtcars$mpg,pred1,pred2,breaks=seq(0.9,1.1,0.01))

SixiangHu/DataMan documentation built on May 9, 2019, 1:48 p.m.