I think Nick is right, and in fact the stata web page says:
"Stata/SE also allows matrices up to 11,000 x 11,000 on computers with
sufficient memory. An implication of this is that Stata/SE can estimate
models with more independent variables (up to 10,998) and estimate
certain panel-data models with larger time-series within panel. "
And
"In addition, Intercooled Stata can have at most 798 right-hand-side
variables in a model."
(== matrices of 800 x 800 is the maximum)
http://www.stata.com/order/options-e.html#difference-sm
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Nick Cox
> Sent: Saturday, March 19, 2005 10:12 AM
> To: [email protected]
> Subject: st: RE: number of regressors
>
>
> There is a general guide to limits
> accessible by
>
> . help limits
>
> The number of regressors is I believe
> bound up with -matsize- in general.
>
> I assert, with some fear of contradiction,
> that the program limit is notional given
> a stronger need for models to be
> scientifically justifiable.
>
> Nick
> [email protected]
>
> Luhang Wang
>
> > Could anyone tell me the built-in variable for the number of
> > regressors in 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/
