Statalist


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

st: Inequality restrictions in ml


From   "Wafa Hakim Orman" <wafa1024@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Inequality restrictions in ml
Date   Sat, 6 Oct 2007 00:21:59 -0500

Hi,

I'm trying to estimate a maximum likelihood model with inequality
restrictions. Basically, I want to constrain a certain linear
combination of a few of the parameters to be positive. (or negative.)

Eg: Suppose the model is

Y = a + bx + cy + dz + e

& I want to restrict: a + b >= 0.

I found this:
http://www.stata.com/support/faqs/stat/intconst.html

in which a procedure to restrict one parameter as positive is
described, but all they appear to be doing is using exp(b) when they
want to restrict b > 0. Seems to me like that's just a non-linear
model rather than restricting the parameter. Or am I missing
something? Is there a better way to do this?

I am using ml so that I can run likelihood ratio tests, but if anyone
knows a method in Stata that does not involve ml, I'd be happy to
know.

Thanks in advance,

Wafa.

-- 
"So be it."
                  --Kurt Vonnegut
*
*   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