View Full Version : can the normal distribution be integrated?

02-03-2002, 10:41 AM
One time I was writing a program for which I needed to plug in the area under the normal distribution. So I tried to "integrate" it to come up with an equation I could plug numbers into to get this area over certain intervals.

But I couldn't integrate it. Not that I would be the one to find it, but is there such an equation?

I suspect that Paul or Renaud could lead us toward the right tool.


02-03-2002, 03:39 PM
Very simple to do actually... excel can do it. Just use the 'normdist' or 'normsdist' function.

02-04-2002, 01:05 AM
But there are very simple and accurate approximations - see Numerical Recipes (available on the web).

02-04-2002, 01:37 AM
Numerical recipes site:


Error function code:


You need to make a change of variable to convert the error function into the normal distribution.

02-04-2002, 09:06 AM

02-04-2002, 09:46 PM
I find it really hard to believe that you do not know how to find how to integrate against a normal distribution. Elroy, for all of your talk do you know the definition of a probability density function?

the integral from negative infinity to t of f(x) times e to the -x squared normalized by 2 pi.

02-05-2002, 01:28 AM