All,
I've just been mucking around with -areg-, which allows me to use
-robust-, something -xtreg, fe- does not allow. A model I've just fitted
with -areg- appears to provide much more sensible estimates and standard
errors than -xtgls- provided (FGLS was finding terms hugely significant
which were insignificant in OLS; and was also reporting Wald values in the
hundreds of 1000s).
However, the description notes in -whelp areg- counsel that:
"-areg- fits a linear regression absorbing one categorical factor; that is,
it fits a fixed-effects model. Note: See the command -xtreg, fe- in help
-xtreg- for an _improved_ version of -areg-." (My emphasis.)
Could somebody tell me how -xtreg, fe- is better than -areg- if it cannot
provide heteroscedasticity-robust standard errors? (Or can it?)
CLIVE NICHOLAS |t: 0(044)191 222 5969
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
*
* 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/
