<>
That would make the problem much simpler, but it would also impose the
constraints that the optimal coefficients on each component of the sum
are equal, which is a much stronger statement than that to be tested
in the user's setup. The suest approach seems sensible.
Kit
On Sep 25, 2009, at 2:33 AM, Nick wrote:
This kind of thing is not my territory at all, but I wondering if
fitting models in which x1+x2+x3 were summed to be a single predictor
would make this easier.
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/