as you suggested, i set all the RHS variables equal,
even the intercept, I got two set of identical
coefficient very similar to what the author came up
with. however I still feel confused on the R-square,
the author only report one R-square. by running sureg,
i got to R-square, and neither of them are identical
to the one reported by the author. is there any way to
constraint the error term?
--- "Rodrigo A. Alfaro" <[email protected]> wrote:
> Another guess, maybe they defined the model as
> twinL-twinH. Rodrigo.
>
> ----- Original Message -----
> From: "Steven Stillman" <[email protected]>
> To: <[email protected]>
> Sent: Friday, October 27, 2006 10:01 PM
> Subject: st: RE: gls result by using sureg
>
>
> 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/
>
> *
> * 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/
>
____________________________________________________________________________________
Access over 1 million songs - Yahoo! Music Unlimited
(http://music.yahoo.com/unlimited)
*
* 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/
