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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Combining inteff, margeff and outreg2


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Combining inteff, margeff and outreg2
Date   Thu, 9 Jun 2011 12:44:54 +0200

On Thu, Jun 9, 2011 at 11:58 AM, Florian Seliger wrote:
> I'm using the margeff and outreg2 command after several probit models:
<snip>
> Is there any way to display the "inteff" marginal effect with outreg2, whereas the other marginal effects are still calculated with margeff?

Not really. This has nothing to do with either -inteff- or -outreg2-
but with the thing you are trying to describe. If you use -inteff-
correctly the result will be a graph showing for each individual its
marginal effect for the interaction term. You will most likely see
that this marginal effect is significantly positive for some,
significantly negative for others and non-significant for the rest.
computing an average marginal effect can thus not do that variation
justice, so we cannot summarize these marginal effects in one number,
so we cannot export it to -outreg2-.

Remember that marginal effects are like a linear model fitted on top
of a non-linear model and models are simplifications of whatever they
describe. Models (of models) are fine as long as they are useful, but
at some point the thing they are describing becomes too complex for
the model to be useful. This is what happens with marginal effects for
interaction effects.

I see two solutions:

1) If all you want to do is present marginal effects, than why not cut
out the middle man and directly estimate a linear probability model?
The probit model is in that case just an unnecessary step between your
data and the results you are after.

2) If the linear probability model is not appropriate for your data
than you can estimate logistic regression models and interpret the
odds ratios, see: M.L. Buis (2010) "Stata tip 87: Interpretation of
interactions in non-linear models", The Stata Journal, 10(2), pp.
305-308.

Hope this helps,
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