--- 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/
