For the underlying programming principles, go straight to a text on
combinatorial algorithms. (Search your univ. library catalogue.)
And then code them in Mata.
The programs I am aware of, -allpossible- and -selectvars- on SSC, are
not good examples of combinatorial programming. They just do what the
author
intended with examples of modest size.
Nick
[email protected]
David Airey
Why not download one of these and view the program code? That's one
nice thing about Stata, you can view code for many of the commands.
On Mar 9, 2008, at 8:13 PM, Raphael Fraser wrote:
> I would like to program all the possible subsets for a regression, say
> regress y x1 x2 x3. How can this be done? I know programs like
> allpossible
> already exist but I am more interested in the "how to" or the
> underlying programming principle.
*
* 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/