Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Dummy variables for pooled regressions


From   "Michael Blasnik" <michael.blasnik@verizon.net>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Dummy variables for pooled regressions
Date   Thu, 12 Jan 2006 07:54:24 -0500

----- Original Message ----- From: "Clive Nicholas" <Clive.Nicholas@newcastle.ac.uk>
To: <statalist@hsphsun2.harvard.edu>
Sent: Thursday, January 12, 2006 2:05 AM
Subject: Re: st: Dummy variables for pooled regressions



Michael Blasnik replied to Yvonne Capstick:

The -egen group- suggestion will provide for fully interacted effects, but
I think that just the main effects are wanted (unless I am guessing wrong
about the somewhat hazy intention). The only way I know of doing this is
to allow areg or xtref, fe to absorb one of the variables and to add
dummies for the other (e.g., using xi).
Like Michael, I too interpreted Yvonne to want two seperate fixed effects
inserted into the -i()- option for her -xtreg, fe- model, which you can't
have. Austin Nichols' rather neat suggestion

. egen g=group(state ind)
. areg y x, absorb(g)
does what I should have figured out: generate a single fixed effect
variable combining both sets of information. But isn't there a problem
with this: namely, the codes for this (in effect) interaction term are not
readily interpretable?

<snip>

I don't think interpretation would be a problem at all -- these are fixed effects, so there is usually no interpretation of their values and indeed they are not actually estimated as coefficients in the model or displayed, so I'm not sure what interpretation issue you would have.

Michael Blasnik
michael.blasnik@verizon.net


*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index