Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: non linear estimation


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: non linear estimation
Date   Wed, 23 Nov 2011 13:44:33 +0000

You can use whatever -nl- supports, which includes the specification of weights. But -nl- is not -regress- PLUS nonlinearity! 

If you want something beyond -nl-, you might need to do much more substantial programming, or approach it differently. This problem is fitting a power function with some constraints and -nl- is not the only way to do it. 

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

Graeml (Dainf) (a.k.a. Rodolfo) 

I´m using a substitutable expression programs (Nonlinear leastsquares 
estimation). The program (below) works very well. However, I would like to 
use robust option or some else that corrects heteroskedastic in nonlinear 
estimation. How can I implement it? Best regards,

program define nlcobPR

            version 9.0

            if "`1'" == "?" {

                global S_1 "a b c"

                global a=0.4

                global b=0.4

                global c=0.2

                global S_2 "Cobb Douglas"

                exit

            }

              replace `1' = $a*((k^$b)*(l^$c)*(t^(1-$b-$c)))

       end

     nl cobPR y if uf=="PR"

     program drop nlcobPR

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index