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 at the end of May, and its replacement, statalist.org is already up and running.


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

st: RE: Newton-Raphson (not concave)


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Newton-Raphson (not concave)
Date   Wed, 14 Apr 2010 11:49:31 +0100

I'd say that while failure to converge is almost always diagnostic of a poor model (or poor data), small struggles in convergence are rarely diagnostic of anything much, especially with complicated models like these. 

But wondering whether a simpler model is preferable is always good science and good statistics. 

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

Dimitrije Tišma

I am running a multi-level random intercept model and after the
adaptive quadrature has converged I have got four consecutive
iterations with (not concave) message. If the model does converge
should I believe my results or should I try another simpler model by
taking out one or two variables? 

*
*   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