I have a Uniform PDF from [-50, 50], I would like to transform it to a Gaussian. The methods that I read up about doing this(like Box Mueller) assume that the uniform distribution is between [0,1). Is there any way to do this for [-50,50]. Is there any translation or scaling that may be done for what I am trying to achieve?
1
-
$\begingroup$ Muller not Mueller. No umlaut and only one e in the original. Open access to original paper at projecteuclid.org/download/pdf_1/euclid.aoms/1177706645 $\endgroup$– Nick CoxCommented Dec 5, 2017 at 9:07
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
You can sample $U \sim \text{Unif}[-50,50]$ and then transform it into $U' = (U+50)/100$, making $U' \sim \text{Unif}[0,1]$, at which point you may apply the Box-Muller transform to the transformed random varible $U'$.
