Hello Statalist-user,
I need to estimate semiparametric as well as parametric models
with time varying covariables in streg and stcox.
I have left-truncated data with multiple records. For my stset command I
made
sure, that the delayed entry is taken into account.
stset month, id(id) fail(du_churn_m) origin(time origin) enter
(month==507)
Estimating the models, I am getting a log pseudo-likelihood for the
Cox-Model
that is negative (-111.533,11) for the Cox-Model.
Indeed for the Exponential and Weibull model the log-pseudo-likelihoods
are positive (+4.766,1459 and 5.043,0699).
I don't know if this is reasonable or there is a problem in the
estimation.
I would really appreciate any assistance with this issue.
Sincerely,
Anouscheh Vahedi
*
* 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/
