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: Wald chi2 test not computed for xtnbreg


From   Nicholas Klein <[email protected]>
To   [email protected]
Subject   st: Wald chi2 test not computed for xtnbreg
Date   Tue, 16 Apr 2013 11:21:42 -0400

Hi all,

I am running a series of –xtnbreg- (random effects) models with a
rather large data set (over 500,000 observations and over 25,000
groups) and running into an odd issue with some of my models.

Some of the models work just fine but others finish but there is no
Wald chi2 or prob>chi2. In both cases, the value is just a period
(dot). The models appear to be converging (log likelihood value is
changing) and this issue arises even when the number of iterations is
not that large (~30).

I am able to get around this problem by using a subsample of my data
and the resulting model coefficients are similar – but I am wondering
if anyone has any idea about why I’m getting such strange Wald chi2
values or have any suggestions for getting around it?

Someone asked a very similar question about three years ago (see
http://www.stata.com/statalist/archive/2010-05/msg01545.html) but I
could not find any replies.


FYI: I’m running the following version of Stata and –xtnbreg- and it
should be up to date.
    Stata/SE 12.1 for Windows (64-bit x86-64)
    Revision 20 Mar 2013
-xtnbreg-
    *! version 1.9.8  15feb2011


--
Nick Klein
[email protected]
PhD Candidate
Edward J. Bloustein School of Planning and Public Policy
Rutgers University

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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