Statalist The Stata Listserver


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

RE: nl and suest [was: st: RE: Chow Test for difference in coeffifients of two regressions]


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: nl and suest [was: st: RE: Chow Test for difference in coeffifients of two regressions]
Date   Fri, 16 Feb 2007 14:10:26 +0100

--- Marisa Miraldo wrote:
> I am interested in estimating a system of equations within which one is non
> linear (and very complex). So I think that what I need is to combine
> nonlinear least squares with the Zellner's seemingly unrelated regression.

I would do this as follows: You can estimate the non-linear
equation with -nl-, as you suggested, and store the 
results with -estimates store-. Estimate the other equation 
with -regress- and estimate that one too with -estimates 
store-. Then combine them with -suest-. See the example 
below. 

However, in the example the standard errors of the nl 
equations get extremely large. Either I made a mistake and 
this procedure is not legal, or the -nl- model doesn't fit at 
all, which isn't surprising since I just chose this model to 
illustrate the steps that need to be taken.

*--- begin example -------
sysuse auto, clear
nl exp3: price mpg
estimates store nl
reg weight foreign price mpg
estimates store ols

suest nl ols
*---- end example --------- 

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology 
Vrije Universiteit Amsterdam 
Boelelaan 1081 
1081 HV Amsterdam 
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434 

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------



*
*   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