# Thread: random number generation with normal distribution

1. ## random number generation with normal distribution

Dear all
Hi,

I seek the formulation of generating a random number whithin for example 8<x<12 but having the normal distribution.

In Excel there is a built in function to do that which is "norm_rnd". But I need the formulation of creating such number

Sincerely
Nasser

2. ## Re: random number generation with normal distribution

Originally Posted by nsvcivil
Dear all
Hi,

I seek the formulation of generating a random number whithin for example 8<x<12 but having the normal distribution.

In Excel there is a built in function to do that which is "norm_rnd". But I need the formulation of creating such number

Sincerely
Nasser
That is because there is no such distribution, the support of the normal distribution is the whole of the reals.

If you want the truncated normal that is a different question.

Also there is no such thing as "the normal distribution", maybe you mean the standard normal, but that seems unlikelt given your restriction on x.

You could try googling for the Box-Muller algorithm.

CB

3. ## Re: random number generation with normal distribution

Dear CaptainBlack

I've the paper of Box and Muller:
Title: "A note on the generation of random normal derivates"
but couldn't understand well!

As I mentioned, by 8<x<12, I mean we have a variance and a mean for an interval. Then we are going to generate a random number from the interval but with a normal distribution

in Excel we have NORM.INV by which we can generate a random number with normal distribution according to a variance and a mean of an interval.

please kindly take a look at this url:
An Introduction to Excel's Normal Distribution Functions

Sincerely

4. ## Re: random number generation with normal distribution

Originally Posted by nsvcivil
As I mentioned, by 8<x<12, I mean we have a variance and a mean for an interval. Then we are going to generate a random number from the interval but with a normal distribution
Please rethink what you really want, because what you are asking, as I said before, is impossible. No normal distribution will produce values restricted to a finite interval.

CB