Re: st: ordered dependent variable

[Maarten Buis <maartenlbuis@gmail.com>]

On Thu, Aug 15, 2013 at 11:39 AM, Nicole Feliciani wrote:
> As you guess my variable is coded from 1 to 6 and I was thinking about
> getting its conditional mean and the use it as the new dependent variable
> with the xtreg command.
> If  I understood rightly this is the tecnique suggested by
> Ferrer-i-Carbonell and van Praag 2004, but I do not know how to implement
> it in Stata.

This is a multi-disciplinary list. The only thing you can be sure
about that we have in common is Stata. This means that references that
are so common in your discipline that you can suffice with only the
names and the year, are likely to be completely unknown on this list.
This is why we ask everybody (in the Statalist FAQ) to give complete
references, just as you would do in a journal.

In principle, there is nothing special you would need to do in Stata:
your dependent variable alreay has the values you want, -xtreg- does
not know that it is actually ordinal and will thus treat it as a
continuous variable. However, especially in more complex models like
-xtreg- I would not do it. The fact that someone else published work
which does it, does not proof that it is OK. In the newest version of
Stata (13) there are now multilevel orderd regression models, that is
where I would look.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/