Statalist The Stata Listserver


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

st: RE: inequality constraints for regression models


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: inequality constraints for regression models
Date   Thu, 6 Apr 2006 17:01:50 +0100

Your constraint is not linear. -constraint- 
should perhaps be called -linearconstraint-,  
but I guess that would not be popular: nevertheless, the
fact that it is restricted to linear constraints
only is mentioned in the help. 

You need to meet this constraint in other 
ways, say by using logit ideas in a reparameterisation. 

Nick 
n.j.cox@durham.ac.uk 

Stephan Brunow
  
> we want to estimate a regression model (Zellners SUR model) with some
> restrictions on parameters and equal explanatory variables. 
> Especially we
> want to say that the parameters have to be between 0 and 1. 
> At the moment we wrote
>  
> Constraint define 1 [eq1]variable1>=0 & [eq1]variable1<=1
> Sureg (yvar1-yvar3 = variable1), cons(1)
>  
> The error message we get is
>  
> 'not possible with test - r(131)'
>  
> What did we wrong? Finally, it turns out that the question 
> is: How to write the constraint?!

*
*   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