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

st: RE: Cox model with shared frailty

From   Dimitris Pavlopoulos <>
To   Garry Anderson <>
Subject   st: RE: Cox model with shared frailty
Date   Fri, 04 Jul 2008 09:54:51 +0200

Sorry, I forgot to mention that I am using STATA 10 MP


Quoting Garry Anderson <>:


What version of Stata are you using?

Cheers, Garry

-----Original Message-----
[] On Behalf Of Dimitris
Sent: Friday, July 04, 2008 5:36 PM
Subject: st: Cox model with shared frailty

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

xi: stcox i.variable, shared(pid)

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

The model has been running for the last 16 hours.

Some time ago, there were discussions in the statalist about it (see,
for example, 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.

Dimitris Pavlopoulos


*   For searches and help try:

dr. Dimitris Pavlopoulos

International Master in Social Policy Analysis (IMPALLA)
University of Leuven
OE CeSO, room 02.27
Parkstraat 45 - bus 3601
3000 Leuven

tel: +32 (0)16 323151


*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index