So unfortunately miest does not do what you want without a little Stata
programming. I don't think however it would be very difficult. The
procedure would go something like this:
1. Estimate model with fixed effects using miest. This procedure allows
you to save the vector of parameter estimates and the variance
covariance matrix (_mib and _miVCE respectively). You need to names
these and save them.
2. Estimate model without fixed effects using miest. Save the parameter
estimates and variance covariance matrix with of course different names
from Step #1.
3. Select the relevant parameters and parts of the variance-covariance
matrix from Step #1 so they match those in Step #2 (that is drop all the
fixed effects if you are estimating fixed effects by the LSDV method, if
not this step shouldn't be necessary as the parameter vectors and varcov
matrices will be of the same dimension and for the same parameters).
4. Conduct Hausman test on your two sets of parameter and varcov
matrices using the standard formulas. This will have to be done using
Stata's matrix functions but the formula is simple so it should not be
too hard.
I know this not quite as easy or quick as you hoped but I think it is
pretty straightforward.
Ken
Carola Herrera wrote:
Good morning,
I have done Multiple imputation with Amelia and I
am doing
the combined analysis of 10 imputed datasets with
"miest". I wonder how to do a Hausman test for
Fixed Effects with "miest". I am running the
regressions with "miest "datasets" reg....and
country dummmies for FE but I can't get Stata to store
the estimates to do the Hausman test afterwards.
Thanks for your suggestions!
Regards,
Carola Herrera
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
--
-----------------------------------------------
Kenneth Scheve
Professor and Director of Graduate Studies
Department of Political Science
Yale University
P.O. Box 208301
New Haven, CT 06520-8301
Tel: +1 203 432 6109
Fax: +1 203 432 6196
Email: kenneth.scheve(a)yale.edu
Web:
http://pantheon.yale.edu/~ks298/index.htm
-----------------------------------------------
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia