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: Dyadic fixed effects and areg, absorb


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Dyadic fixed effects and areg, absorb
Date   Fri, 15 Apr 2011 15:54:27 +0100

It starts sounding like a statistical equivalent of a rather empty
model. If I say, interactions between countries over time depend on
precisely which countries and precisely when, the reaction to those
words is, Well, yes. If you say, and here are thousands of parameters
to summarize that, how is that scientifically or practically
interesting or useful?

I don't know what your overall objective is. Are you a student and you
want to pass an examination or to get  a degree? Are you a researcher
and you want a publishable paper? Either way, as you ask for
reactions, mine is that I would expect anyone in your friend to be
more interested in a much more parsimonious model.

On Fri, Apr 15, 2011 at 3:31 PM, emanuele mazzini
<madsoenistata@gmail.com> wrote:
> Dear all Stata users,
> I am trying to work out a regression with an unbalanced bilateral
> panel dataset in which I need to include both dyadic fixed effects and
> year fixed effects. Since my dataset is very huge (it includes more
> than 600,000 observations) and I cannot generate a dummy for the
> variable dyads (which is the unique dyads identifier), I thought that
> I can accomplish this by using the following command:
>
> areg...., absorb(dyads)
>
> by including only the dummies that refer to the years to take into
> account the year fixed effects, while absorb(dyads) is supposed to
> take into account the dyadic fixed effects.
>
> I am not looking really for an help, but more for an advice: what do
> you think about this solution to my problem (i.e. to the impossibility
> to generate more than 15,000 dummies)? Do you think this may work?
>
> Thanks to all of you in advance for your time and consideration,
*
*   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