Skip to content

Tippett plot

2 messages · Spiros Stavropoulos (DT), Jim Lemon

#
Spiros Stavropoulos (DT) wrote:
Hi Spiros,
You can get pretty much what you want with:

# make up probability and likelihood data
p1<-log(cumsum(abs(rnorm(20))))
l1<-rev(rescale(cumsum(abs(rnorm(20))),c(0,1)))
# plot it in stairstep mode
plot(p1,l1,type="s")
# make up some more data and add another line
p2<-log(cumsum(abs(rnorm(20))))
l2<-rev(rescale(cumsum(abs(rnorm(20))),c(0,1)))
lines(p2,l2,type="s",lty=2)

Jim