From: dtprice@interaccess.com (David T. Price)
Subject: Re: erf/cumnorm function algorithm?
Date: Fri, 30 Apr 1999 23:11:06 GMT
Newsgroups: sci.math.num-analysis
Keywords: algorithms to compute erf

jordan.samuels@wdr.com wrote:

>
>
>Does anyone know the algorithm for erf(3M) on Solaris or how to find
>out what it is?

Check out the work of W. J. Cody cited in section 4.2 of "Numerical
Evaluation of Special Functions" at http://math.nist.gov/nesf/ for the
standard alogrithms for these functions.

Dave Price
==============================================================================

From: "Alan Miller" <amiller @ vic.bigpond.net.au>
Subject: Re: erf/cumnorm function algorithm?
Date: Sat, 1 May 1999 09:55:56 +1000
Newsgroups: sci.math.num-analysis

For the normal distribution function, look at Applied Statistics
algorithm AS 66 which you can download from the apstat directory
at statlib.   You can also obtain it from my web site.

N.B. The cumulative normal is integrated from minus infinity to x,
not from zero.

statlib:   http://lib.stat.cmu.edu  or ftp from: lib.stat.cmu.edu

--
Alan Miller, Retired Scientist (Statistician)
CSIRO Mathematical & Information Sciences
Alan.Miller -at- vic.cmis.csiro.au
http://www.ozemail.com.au/~milleraj
David T. Price wrote in message <372a3786.41364796@nntp.interaccess.com>...
[previous article quoted -- djr]
==============================================================================

From: dtprice@interaccess.com (David T. Price)
Subject: Re: erf/cumnorm function algorithm?
Date: Sat, 01 May 1999 02:12:17 GMT
Newsgroups: sci.math.num-analysis

dtprice@interaccess.com (David T. Price) wrote:
[first article quoted -- djr]

The code for Cody's algorithm is available at
http://www.netlib.org/specfun/erf.

Dave