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:about the fixed-effects model


From   "Jing Zhou" <jing.zhou@rmit.edu.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st:about the fixed-effects model
Date   Tue, 31 Aug 2010 18:01:03 +1000

Dear Maarten,

Thanks for your reply. My dataset comprises firms observed overtime. the time-invariant variable is computed at the province level. Can I use the FE model to estimate this time-invariant variable? is that what you refer to as "your grouping variable did not identify the groups in the way you think it should be done"?

Thanks!

Jing

>>> Maarten buis <maartenbuis@yahoo.co.uk> 31/08/2010 5:04 pm >>>
--- On Tue, 31/8/10, Jing Zhou wrote:
> It is suggested that the FE model would not be appropriate
> to estimate the time-invariant variable, as such variable
> will be excluded by differencing the values. however, in my
> FE model, the time-invariant variable does not disappear and
> may be even significant. could you please advise me why it
> happens and whether I can trust this result?

There is nothing magical about the variable time that says that
time-invariant variables must dissappear in fixed effects 
regressions, but in many instances we use fixed effects 
regressions in cases where individuals/firms/countries/...
are observed over multiple time points, in which case the
time invariant variables should dissappear. 

If your dataset also is of the form individuals observed over 
time, then it means that something went wrong: either you think 
you estimated a fixed effects model but you did not, or your 
grouping variable did not identify the groups in the way you 
think it should be done, or your time invariant variable is not 
time invariant. 

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/


*
*   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