Just a guess, but perhaps they constrained the coefficients for each twin to
be identical. This can be done using the multiple-equation syntax of sureg
and a cross-equation constraint for each RHS variable.
Cheers,
Steve
-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of hai guo
Sent: Saturday, October 28, 2006 10:10 AM
To: [email protected]
Subject: st: gls result by using sureg
Ashenfelter and Krueger 's paper in 1994 on the
estimates of the economic return to schooling from a
new sample of twins reported a GLS result. I am trying
to duplicate the result, however I don't know how. in
the dataset, they have wage, education, age, agesqu,
male and white for twin H and twin L.as the result
table indicate that the sample size for the gls result
is 298 combining the twin H and L. in other words, it
has 149 Twin H and 149 Twin L. in the paper they
metioned that they apply the
seemingly-urelated-regression method due to ARnold
zellner to fit the GLS. so I tried to used
sureg(lhrwagel educl age agesq malel whitel) (lhrwageh
educh age agesq maleh whiteh) . however i got two sets
of coefficient with sample size 149 for each. I am
wondering how can i get only one set of coefficient
with sample size of 298. thanks
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* 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/
