Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: ZOIB procedure


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: ZOIB procedure
Date   Tue, 20 Sep 2011 11:41:39 +0200

--- On Tue, Sep 20, 2011 at 11:20 AM, Prerna S wrote:
> As a follow-up question, the statistical significance of the marginal
> effects differs from the statistical significance of the zoib
> regression estimates. How does one explain this difference? And in
> explaining results, does one choose zoib results or marginal effects?

I would go for marginal effects. You can interpret exponentiated
coefficients as odds ratios (for being an exact 0 or an exact 1,
depending on the equation) and relative proportion ratios given one is
not an exact 0 or 1 (see -help betafit- for an example on relative
proportion ratios). The problem is that these are conditional effects
while you probably want the effect on the unconditional expected
proportion, and these ratios are non-standard so they need a lot of
explaining.

I can think of two reasons why the statistical significance differs,
and they are all a variation on the point that they represent
different null-hypotheses: 1) The parameters represent relative
effects (ratios), while marginal effects represent absolute effects
(differences). 2) The marginal effects combine the effects on all two
or three equations of the model.

-- Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index