Re: st: Modelling two binary outcomes that are not mutually exclusive

[Ron�n Conroy <rconroy@rcsi.ie>]

SamL wrote:

> I don't understand why you wouldn't use a bivariate probit model.  It is
> my understanding that this model is designed for 2 binary outcomes.  The
> error terms in the two equations are allowed to be correlated (the
> correlation is estimated).  Many people use this with one equation 
> being a
> "selection" equation, but that is not required.  If I understand your
> issue correctly, this would be an appropriate model--not multinomial
> logit.
I want a relative risk ratio - that is, I want the comparison group for 
the two models to be the group who have no diagnosis. This isn't what 
bivariate probit models do, is it? (Please correct me if I'm wrong - I'm 
not familiar with them)

--

Ronan M Conroy (rconroy@rcsi.ie) 
Senior Lecturer in Biostatistics 
Royal College of Surgeons 
Dublin 2, Ireland 
+353 1 402 2431 (fax 2764) 
-------------------- 
Just say no to drug reps 
http://www.nofreelunch.org/

*
*   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/