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: predict(pu0) after clogit


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: predict(pu0) after clogit
Date   Tue, 27 Mar 2012 12:14:23 +0200

On Tue, Mar 27, 2012 at 11:55 AM, Ori Katz <ori.k66@gmail.com> wrote:
> What do you think about using "margins, dydx(*) predict(pu0)" in my
> case, after the clogit?
>
> In the results table it says "Pr(choice1|fixed effect is 0)", but I
> don't understand what does it means by "fixed effect is 0". I don't
> have fixed effects in the model, only majors income and interaction
> between major income and the ethnic origin of the individuals.

You have fixed effects: that is what -clogit- is giving you. If you do
not want fixed effects than -clogit- is the wrong command.

I don't think that the - predict(pu0)- option in -margins- is what you
want as it is a rather weird hybrid between average marginal effects
and marginal effects at average values of explanatory variables. The
fixed effects are implemented such that the value 0 represents the
average (which is why you have a constant in -clogit-). So your
margins command gives you marginal effects averaged over your observed
variable but at the same time fixing the fixed effects at its average.

The best thing you can do is work with odds, odds ratios, and ratios
of odds ratios.

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