[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sas-l
Subject:    Re: log reg w/ multiple dependent variables
From:       Citam <citam.sasl () GMAIL ! COM>
Date:       2008-04-30 15:23:59
Message-ID: 963e8d4c0804300823l38940b8ci4e2b7be4d4ed8fb7 () mail ! gmail ! com
[Download RAW message or body]

On 4/29/08, Rachel Fusco <rafusco@msn.com> wrote:
> I want to run a logistic regression with one independent variable and
> multiple dependent variables. Is this possible? Thank you.

Sure.  I can provide an example.  In genetics, one might be interested
in a bivariate analysis controlling for genotype at one locus.
Although this would be a minimally bare model for which I might assume
the sample is restriced to one sex, it is possible.  One possibility
is that association does not necessary indicate causation.  You need
to consider whether you exposure variable affects (causes)your
outcomes and whether any outcome variable is intermediate to another.

Now the question might be whether multivariate logistic regression is
possible in SAS.  I would start by perusing the GENMOD procedure.
[prev in list] [next in list] [prev in thread] [next in thread]