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

# st: not concave log pseudolik in gnbreg

 From "paolo.dibetta@fastwebnet.it" To statalist@hsphsun2.harvard.edu Subject st: not concave log pseudolik in gnbreg Date Thu, 10 Jan 2013 23:20:06 +0100 (CET)

```I am running a model in -gnbreg-

The log takes many iterations and the final part reads as follows:

Iteration 1320:log pseudolikelihood =  -399.5054
Iteration 1321:log pseudolikelihood =  -398.9085
Iteration 1322:log pseudolikelihood = -398.79423
Iteration 1323:log pseudolikelihood = -398.77882
Iteration 1324:log pseudolikelihood = -398.74555  (not concave)
Iteration 1325:log pseudolikelihood = -398.74486
Iteration 1326:log pseudolikelihood = -398.74102
Iteration 1327:log pseudolikelihood = -398.74077  (not concave)
Iteration 1328:log pseudolikelihood = -398.74077

Generalized negative binomial regression          Number of obs   =        341
Wald chi2(18)  =          .
Prob > chi2     =          .
Log pseudolikelihood = -398.74077                 Pseudo R2       =     0.4011
//(table of results omitted)

What is disturbing is that
1. it says "(not concave)" and then replicates the same value;
2. it does not give the Wald test.

I know that the true problem is when the last iteration says
"(not concave)" and then the results come out,
so I do not know if I can trust the results in this case.

What is going on? I wonder if I can use the results.
(The results of the overall model make sense, by the way. It is the best I have,
with the lowest AIC and BIC. High multicollinearity is a problem, but I need to
keep interaction terms.)

Thank you very much in advance for any help, much appreciated.

Best,
Paolo Di Betta
University of Palermo, Italy

*
*   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/
```