Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: question regarding nl


From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: question regarding nl
Date   Wed, 29 Oct 2003 13:01:56 -0000

-nl- is indicating, inarticulately, that there is a
problem. Perhaps you ended up in a strange corner of 
parameter space. Try different starting values. Be
reluctant to believe a fit without reasonable and
indeed finite standard errors. 

0 for missing is, I think, a historical hangover
from days in which matrices could not include 
missing values, so I support your surmise. 

Nick 
[email protected] 

Harald Badinger

> does anyone know how to interpret standard errors of 0 for part of 
> the coefficients in the nonlinear least sqaures output.
> 
> Actually, the estimation output does not report standard errors, t-
> and p- values for some of the parameters; however, looking at the 
> standard errors using the _se[]command or the e(V) command, the 
> values displayed are "0".
> (I guess the opposite is true: that the standard errors are 
> extremely large).
> 

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index