st: re: comparison two SUR models

[Christopher Baum <kit.baum@bc.edu>]

<>
I'm using seemingly unrelated regression (SUR) technique to estimate
four models (equations). I have imposed symmetry in this model (some
coefficients amog different equations are the same). I have also run
the same 4 equations under the same approach without imposing
symmetry. I want to compare both models to see which one is more
appropriate.

Instead of imposing the constraints and estimating the SUR system with symmetry, estimate the unrestricted system
and use -test- to jointly specify all of the symmetry restrictions. That will give you an F test (with numerator df equal to the number of restrictions) that will allow you to consider the null of symmetry.

There is a worked example of this in section 9.4 of IMEUS.

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/