-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On
Behalf Of Glen Barnett
Sent: Tuesday, July 23, 2002 10:44 PM
To: [EMAIL PROTECTED]
Subject: Re: Integrating Standard Normal pdf
Gordon Kenyon <[EMAIL PROTECTED]> wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> In quantitative methods text-books for the social and behavioral
> sciences one constantly sees the equation which describes the pdf for
> both normal distributions in general and the standard normal
> distribution. I have never seen any basic warning that this
> "description" of the standard normal distribution is only the density
> function and not a computing formula of any kind.
Of course it's a computing formula. You use it any time you need
the density!
> Anyway, integrating the normal distribution function sure is a bear,
> huh?
If you mean the incomplete integral, well, yes, you can't do it
analytically. But the whole density is relatively easy to integrate
once someone shows you a way to do it.
But numerical approximation of the integral is pretty easy.
(For hand work, I just use simple quadrature rules,
but taken from a point I already have memorised.
Even Simpson's rule is often good enough.)
Glen
-------------------------------------------------------
The standard infinite series is also good and accurate out to about 10
sigma. It can be programmed in any language.
DAHeiser
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
