As I wanted to simulate truncated normals in a hurry, I coded the inverse cdf approach:
[sourcecode language=”r” gutter=”false”]
truncnorm=function(a,b,mu,sigma){
u=runif(1)
u=qnorm(pnorm((a-mu)/sigma)*(1-u)+u*pnorm((b-mu)/sigma))
return(mu+sigma*u)
}
[/sourcecode]
instead of using my own accept-reject algorithm. Poor shortcut as the method fails when a and b are too far from μ
[sourcecode language=”r” gutter=”false”]
> truncnorm(1,2,3,4)
[1] -0.4912926
> truncnorm(1,2,13,1)
[1] Inf
[/sourcecode]
So I introduced a control (and ended up wasting more time than if I had used my optimised accept-reject version!)
[sourcecode language=”r” gutter=”false”]
truncnorm=function(a,b,mu,sigma){
u=runif(1)
if (pnorm((b-mu)/sigma)-pnorm((a-mu)/sigma)>0){
u=qnorm(pnorm((a-mu)/sigma)*(1-u)+u*pnorm((b-mu)/sigma))
}else{
u=-qnorm(pnorm(-(a-mu)/sigma)*(1-u)-u*pnorm(-(b-mu)/sigma))}
return(mu+sigma*u)
}
[/sourcecode]
As shown by the above, it works, even when a=1, b=2 and μ=20. However, this eventually collapses as well and I ended up installing the msm R package that includes rtnorm, an R function running my accept-reject version. (This package was written by Chris Jackson from the MRC Unit in Cambridge.)
