Hi!
I am using the Jenkins pgmhaz procedure to estimate a discrete time non-parametric proportional hazard model.
I would like to ask about the estimation of the second model. During the iteration process I have received different messages: unproductive step attempted and nonconcave function encountered, but in the end the convergence has been achieved. However, when I run a slightly different specification I now receive the following error message: pgm_ll does not compute a continuos nonconstant function/ could not calculate numerical derivatives. Does this imply that there was no convergence? What is the best way to proceed in such case?
Thank you in advance,
Laura Vartia
*
* 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/
