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

Re: st: Fixed Effects estimation with time-invariant variables


From   David Jacobs <[email protected]>
To   [email protected]
Subject   Re: st: Fixed Effects estimation with time-invariant variables
Date   Wed, 27 Jul 2005 12:44:12 -0400

It gives you a period specific estimate of the time-invariant variables effects. Such estimates can be instructive. For example, by not forcing the coefficients on the explanatory variable to be the same in different periods, you may find that a variable you thought did not matter does. Such an analysis also can tell you if the effect in question is increasing or decreasing in strength.

Dave Jacobs

At 06:44 AM 7/27/2005 -0400, you wrote:

I do not understand why one would do that? Rather, what purpose would it serve?
Best,
Alice


From: David Jacobs <[email protected]>
Reply-To: [email protected]
To: [email protected]
Subject: Re: st: Fixed Effects estimation with time-invariant variables
Date: Tue, 26 Jul 2005 16:08:11 -0400

If you create interaction terms between between three of the four possible dummies computed on your periods and multiply these dummies by one of your time invariant explanatory variables, your fixed-effects equation will produce estimates for these interacted "ex" time-invariant effects. See Charles Halaby's piece on panel estimation in an Annual Review of Sociology about two to four years ago.

Dave Jacobs

At 01:56 PM 7/26/2005 +0100, you wrote:

Fixed Effects estimation with time-invariant variables

I have a problem and I wonder whether anyone can help. I have a panel
dataset of 32 Sub-Saharan African countries and four time periods
(N=32, T=4). I am looking at the determinants of aid for these
countries. Given that they are not randomly selected countries, I am
using Fixed Effects. However, I have a few time-invariant variables
that are important but that get dropped because of using FE.  Is there
any way of obtaining estimates for these time-invariant variables and
still use FE?  Also, given the data I'm using, is there any
justification for using Random Effects?

Any comments would be much appreciated.

Thanks.
Joana

*
*   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/

*
*   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/
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/

*
* 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/

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index