st: Cox model with shared frailty

[Dimitris Pavlopoulos <Dimitris.Pavlopoulos@soc.kuleuven.be>]

Dear Statalisters,

I would like to ask for your help in running a cox model with shared  
frailty on unemployment duration. I have organized my data in a spell  
file with in total 7,401 spells. I tried a very simple version of the  
model:

xi: stcox i.variable, shared(pid)

where pid is the person identification number and 'variable' has 4 categories.

The model has been running for the last 16 hours.

Some time ago, there were discussions in the statalist about it (see,  
for example,
http://www.stata.com/statalist/archive/2004-01/msg00728.html). I  
understand from these discussions that the cox model with frailty runs  
extremely slow due to the exact calculation of the partial likelihood  
from STATA. As these questions were posted a few years ago, my  
question is whether there has been developed a way to speed up the  
process. Or would the only way be to drop the cox model and try a  
parametric one?

Thank you in advance.

best,
Dimitris Pavlopoulos


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

*
*   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/