Re: st: Re: comparing OR (coefficients) across equations

["R.E. De Hoyos" <redeho@hotmail.com>]

Ricardo,

I guess you have three groups in your sample: cancerA, cancerB and other. 
You can fit an -mlogit- with these three categories to see if the 
independent variables have a different effect upon the likelihood of showing 
a particular disease.

Once you have the coefficients into a single -mlogit- it is relatively easy 
to test Ho. of equality in parameters.

I hope this helps,

Rafa


----- Original Message ----- 
From: "Ricardo Ovaldia" <ovaldia@yahoo.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Monday, November 01, 2004 8:48 PM
Subject: Re: st: Re: comparing OR (coefficients) across equations



> --- "R.E. De Hoyos" <redeho@hotmail.com> wrote:
>
> > Ricardo,
> >
> > According to the Statalist archives you have asked
> > this question before
> > (30/Aug/04) and received several informed replies.
> http://www.stata.com/statalist/archive/2004-08/msg00870.html
> > Did I miss something?
> The difference is that in that situation I wanted to
> compare coefficients computed from the same model(same
> outcome & IV)applied to two different populations. For
> my current question I have the same independent
> variables but the outcome variables is different. Two
> different cancers, same control group.
>
> I thing this is different, but I am not sure. Also, I
> was wondering if -mhodds- would work in this case.
> Note that I have both continuous and discrete
> covariates.
>
> Thanks,
> Ricardo.
>
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Check out the new Yahoo! Front Page.
> www.yahoo.com
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/