Dear statalisters,
I have a question which is half Stata / half statistics:
in their famous twin study on returns to education, Ashenfelter and Krueger
(1994, AER) report GLS estimates and say on p. 1164 that these come from the
seemingly-unrelated regression method due to Zellner (1962), so this should
be Stata's sureg. Now, specifying
sureg (lhrwagel educl educh age agesq whitel malel)
(lhrwageh educh educl age agesq whiteh maleh) if t==1
where "l" stands for the first twin and "h" for the second twin, I obviously
get different coefficients on the two equations of the system. (Note that
there are 149 twin pairs, i.e. 298 individuals, so 149 "first twins" which go
in the first equation and 149 "second twins" which go in the second equation)
Yet, Ashenfelter and Krueger report only one set of coefficient estimates
which does not seem to be the average of the two sets of sureg coefficients
either.
Any idea of how they get their results in table 3, column 2 ?
Thanks, Sascha
--
+++ GMX - Mail, Messaging & more http://www.gmx.net +++
Bitte l�cheln! Fotogalerie online mit GMX ohne eigene Homepage!
*
* 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/
