Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: RE: RE: RE: RE: NL SUR


From   "Brian P. Poi" <brian@poiholdings.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: RE: RE: NL SUR
Date   Mon, 22 Nov 2010 18:33:04 -0500

-----Original Message-----
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of A.Silva
Sent: Monday, November 22, 2010 9:33 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: NL SUR

Dear all,
I am trying to estimate a NL SUR (AIDS). I have 6 expenditure groups (so, I
use 5 equations).
However, I get the following error:

nlsur aids @ w1 w2 w3 w4 w5 lnp1 lnp2 lnp3 lnp4 lnp5 lnp6 lnpx, parameters
(a1 a2 a3
> a4 a5 b1 b2 b3 b4 b5 g11 g12 g13 g14 g15 g22 g23 g24 g25 g33 g34 g35 
> g44
g45 g55) neq
> (5) ifgnls
(obs = 93)

Calculating NLS estimates...
could not evaluate equation 1
starting values invalid or some RHS variables have missing values r(480);

end of do-file

r(480);

. sum w1 w2 w3 w4 w5

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
          w1 |        93    .1100006     .009555   .0885841   .1273791
          w2 |        93    .0358503    .0024138   .0261827   .0412181
          w3 |        93    .3097675    .0091194   .2870499   .3364181
          w4 |        93    .1936698    .0147691   .1689364   .2275145
          w5 |        93     .085791    .0040142   .0737699   .0976536

-------------

Aside from any syntax errors or other bugs in your code, the real
fundamental problem here is the small number of observations.  Typically
these demand systems are fitted using thousands of observations, and in the
past I've seen people have convergence problems when they are trying to fit
a demand system that's just too complicated for the amount of data
available.  Given that you only have 93 observations and the expenditure
shares don't appear to have all that much variation, I'd be surprised if you
could really estimate all 25 parameters with any degree of precision.


Just something to think about.

   -- Brian Poi
   -- brian@poiholdings.com


*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index