Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: convergence problem


From   "Carlo Lazzaro" <carlo.lazzaro@tin.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: convergence problem
Date   Mon, 10 Sep 2007 15:50:14 +0200

Dear Statalisters,

fittine a generalized gamma model with Stata 9.2/SE for testing via K=1 test
the appropriateness of the Weibull parametric survival model, I get the
following lines and error (not convergence).


. streg DMD, dist(gamma) technique(nr)

         failure _d:  MS_Conv == 1
   analysis time _t:  risk_time
                 id:  id

Fitting constant-only model:

Iteration 0:   log likelihood = -218277.41  (not concave)
Iteration 1:   log likelihood = -176430.05  
Iteration 2:   log likelihood = -154374.58  (not concave)
Iteration 3:   log likelihood = -129998.81  
Iteration 4:   log likelihood = -109341.76  (not concave)
Iteration 5:   log likelihood = -98110.045  
Iteration 6:   log likelihood = -90070.512  
Iteration 7:   log likelihood = -85689.478  
Iteration 8:   log likelihood = -84370.382  
discontinuous region encountered
cannot compute an improvement
r(430);

Is there any other way for working this problem around and checking for the
appropriateness of Weibull model?

Thanks a lot for your kindness and for your time.

Kind Regards,

Carlo


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index