On Jan 15, 2004, at 2:33 AM, Apostolos wrote:
I want to evaluate a system of equations using SUR. I did not program
the
equations directly but rather I used the menu system of v. 8. I
discovered
that it is very, very slow to converge and I also found it quite
irritating
that it paused to inform me about the convergence criterion. What am I
doing wrong?
Beyond using menus? :)
Seriously, the "isure" option must be specified for -sureg- to iterate
at all. SURE is a linear estimator, and does not need to do any
iteration to generate its results. You may choose to iterate the SURE
estiimator, but the default behavior of sureg does not do so. If there
are (numerically) different explanatory variables in your equations,
iteration may improve the efficiency of the estiimates, but from a
large-sample perspective one turn of the crank is sufficient.
Kit
*
* 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/
