Date:         Thu, 23 Jun 2005 13:57:43 -0400
Reply-To:     Peter Flom <flom@NDRI.ORG>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Flom <flom@NDRI.ORG>
Subject:      Re: Exact confidence interval for the difference
Comments: To: cassell.david@EPAMAIL.EPA.GOV
Content-Type: text/plain; charset=US-ASCII

There's another, more recent article by Agresti and Coull that is
apropos

@ARTICLE{,
   author = {Alan Agresti and B. A. Coull},
   title = {Approximate is better than "exact" for interval estimation
of binomial proportions},
   journal = {American Statistician},
   year = {1998},
   volume = {52},
   number = {},
   pages = {119-126},
   month = {},
}

HTH

Peter

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)



>>> "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV> 6/23/2005
1:18:00 PM >>>
Bob_Abelson@HGSI.COM wrote:
> One of our statisticians wants to know:
>
> Does anyone know if SAS or R can compute an exact confidence
interval
for
> the difference
> in 2 binomial proportions?...the only package that I have ever seen
do
> this calculation is StatXact...

SAS doesn't do an exact CI for the difference in 2 binomial
proportions.

This is fine by me.  If you read Alan Agresti's 1992 paper "A Survey
of Exact Inference for Contingency Tables", you'll see that he finds
there are problems with defining an 'exact' test in this case.

HTH,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician