Vumani,
This is a topic that has come up before on the list. See my posting at
http://www.biostat.wustl.edu/s-news/s-news-archive/199906/msg00205.html
(June 24, 1999) for some references. I have found Gange's method to be the
most flexible.
Chris Bilder
Christopher R. Bilder, Ph.D.
Assistant Professor
Oklahoma State University
Department of Statistics
301G MSCS Bldg.
Stillwater, OK 74078
Phone: (405) 744-5684
Fax: (405) 744-3533
E-mail: bilder@okstate.edu or chris@chrisbilder.com
Website: www.chrisbilder.com
GO BIG RED!
-----Original Message-----
From:
Sent: None
Subject:
s-news-owner@lists.biostat.wustl.edu Thu Jul 5 06:44:49 2001
Received: from hotmail.com (f220.law11.hotmail.com [64.4.17.220])
by wubios.wustl.edu (8.9.3/8.9.3) with ESMTP id GAA06424
for <s-news@lists.biostat.wustl.edu>; Thu, 5 Jul 2001 06:44:22 -0500
(CDT)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
Thu, 5 Jul 2001 04:43:50 -0700
Received: from 131.211.95.179 by lw11fd.law11.hotmail.msn.com with HTTP;
Thu, 05 Jul 2001 11:43:50 GMT
X-Originating-IP: [131.211.95.179]
From: "Vumani Dlamini" <dvumani@hotmail.com>
To: s-news@lists.biostat.wustl.edu
Subject: sampling correlated binary variables
Date: Thu, 05 Jul 2001 11:43:50 -0000
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F220IZS2TRqf69aZbLA000071f5@hotmail.com>
X-OriginalArrivalTime: 05 Jul 2001 11:43:50.0737 (UTC)
FILETIME=[C2849410:01C10547]
X-Archive-Number: 200107/25
X-Sequence-Number: 2366
I would like to sample correlated binary variables, which have a different
probability of success.
"rbinom" gives me independent variables, and sampling the first covariate,
and then using it as a covariate of the logit of the probability of the
second covariate doesn't look elegant.
Any help is welcome.
Vumani Dlamini
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
|