Date:         Sun, 23 Mar 1997 14:19:54 GMT
Reply-To:     Michael Friendly <friendly@HOTSPUR.PSYCH.YORKU.CA>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Michael Friendly <friendly@HOTSPUR.PSYCH.YORKU.CA>
Organization: York University, Ontario, Canada
Subject:      Re: Bootstraping in SAS

In article <3331F38B.16F@hjs1.iae.tu-berlin.de> Joseph Ogutu
<ogutu@hjs1.iae.tu-berlin.de> writes:
|Walter Johnston wrote:
|
|> Yes, it is possible - and not particularly difficult. But what are
|> you wanting to bootstrap?
|>
|> Walter Johnston, Ph.D. (Statistics)
|
 
There have been several fairly general bootstrapping programs written
in SAS. The simplest I've seen is in the JSE SAS file archive at
gopher://jse.stat.ncsu.edu/11/software/sas
 
Pat Thompson had another in last year's SUGI proceedings,  Charlie
Hallahan had a SAS/IML version in the Sage book edited by John Fox and
Bob Stine, Statistical Computing Environments (not the exact title).
 
The most general one is the JACKBOOT.SAS program from SAS Inst, available
in the techsup/download/stat directory (FTP: ftp.sas.com).   You supply
a %analyze macro which does whatever analysis you want, and it
bootstraps or jackknifes the analysis over the data.
 
--
Michael Friendly     Internet: friendly@hotspur.psych.yorku.ca (NeXTmail OK)
Psychology Dept
York University      Voice: 416 736-5118  Fax: 416 736-5814
4700 Keele Street    http://www.math.yorku.ca/SCS/friendly.html
Toronto, ONT  M3J 1P3 CANADA