On #1, the likelihood depends on the model
and the data. Therefore, you must
change one or both of those to make
any difference here.
On #2, the estimates may look identical
to the precision printed, but -ml-
may still be climbing, or trying to climb.
Nick
[email protected]
Prathiba Natesan
> I am evaluating a polytomous IRT model in Stata using GLLAMM and
> frequently encounter the following errors. Can someone help?
> 1. Often I get a msg that says, flat or discontinuous region
> encountered. What can I do to fix this?
> 2. Sometimes, the log likelihood produces identical estimates as the
> previos estimate and still continues to run. Does this mean that it
> has gotten into an unconditional looping or does it mean that there is
> some problem with the data?
*
* 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/
