Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: mvprobit


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: mvprobit
Date   Thu, 28 Feb 2008 15:14:38 +0000 (GMT)

--- Simon Moore <simoncmoore@gmail.com> wrote:
> Is there an accepted criteria for use of mvprobit over two or more 
> separate probit equations?  My only lead so far, from the published 
> literature using mvprobit, tells me that mvprobit is best if outcome 
> variables are correlated but not so correlated that dependant
> variables should be collapsed into a single variable.

It is not so much that the depedent variables are correlated, but that
the error terms are correlated. I think of it as there is some
unobserved variable that influences all dependent variables. Given my
simplistic mind I need a concrete variable, I often think about it in
terms of smart respondents are more likely to be succesful interms of
dependent variable 1 and dependent variable 2 (assuming you haven't
measured IQ or something like that). In other words the criterium is
mostly theoretical.

Hope this helps,
Maarten


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      __________________________________________________________
Sent from Yahoo! Mail.
A Smarter Inbox. http://uk.docs.yahoo.com/nowyoucan.html
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index