Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Constraint nl estimation


From   "Joseph Coveney" <jcoveney@bigplanet.com>
To   "Statalist" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Constraint nl estimation
Date   Tue, 4 Mar 2008 12:13:54 +0900

Vladimir Vakhitov wrote:

I run a nl estimation on the function a part of which is
(gX1+(1-g)X2). Naturally I would like to constraint g to the interval
(0,1), not including the boundaries. It looks like there is no
"constraint" function for nl procedure. Any ideas?

--------------------------------------------------------------------------------

You can use Stata's inverse-logit transformation to map real numbers into
(0, 1).  Use the -invlogit()- function in your -nl- model specification.
Somewhere I saw a list of handy nonlinear transformations, ln for strictly
positive, hyperbolic arc tangent for -1 to +1, etc.

Also, if you're interested in linear constraints, you can use -ml- for
nonlinear model fitting in lieu of -nl-.  To see how, take a look at the FAQ
www.stata.com/support/faqs/stat/nl_ml8.html .

Joseph Coveney


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index