Skip to content

Instantly share code, notes, and snippets.

@clayford
Last active December 30, 2015 14:29
Show Gist options
  • Save clayford/7842136 to your computer and use it in GitHub Desktop.
Save clayford/7842136 to your computer and use it in GitHub Desktop.
Plotting confidence intervals to see which contain true mean.
# create a blank plot
plot(x=c(10,20),y=c(0,40), type="n")
abline(v=15,lty=2)
# plot line segments to represent confidence intervals
# add red lines for CI's that do not capture true mean of 15
# add dots for means
for(i in 1:40){
x <- rnorm(30,15,2)
res <- t.test(x,mu=15)
if(res$conf.int[1]>15 | res$conf.int[2]<15){
segments(x0=res$conf.int[1],y0=i,x1=res$conf.int[2],y1=i,col="red")
points(x=res$estimate, y=i, pch=19, col="red")
} else {
segments(x0=res$conf.int[1],y0=i,x1=res$conf.int[2],y1=i)
points(x=res$estimate, y=i, pch=19)
}
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment