Skip to content

Instantly share code, notes, and snippets.

@lschwetlick
Created January 6, 2021 17:20
Show Gist options
  • Save lschwetlick/562366af0197205894beb80961b1dd94 to your computer and use it in GitHub Desktop.
Save lschwetlick/562366af0197205894beb80961b1dd94 to your computer and use it in GitHub Desktop.
How to generate a reciprobit plot (used in LATER models) in R
# We want to generate a reciprobit plot like it is used in the context of LATER models.
# Here, we generate normally distributed data. In the reciprobit plot that should appear linear
rm(list=ls())
library(ggplot2)
N = 1000
f = rnorm(N,mean=300,sd=50)
x = sort(1/f,decreasing=T)
y = (1:N)/N
d = data.frame(x=x,y=y)
l = seq(from=1,to=0.00,by=-0.001)
s = round(1/l)
p = ggplot(d,aes(x=x,y=y)) +
geom_point() +
geom_line() +
scale_x_continuous(trans="reciprocal",breaks=l,labels=s) +
scale_y_continuous(trans="probit")
print(p)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment