Cox regression - time variable

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Posts: 2
Joined: Fri Jul 12, 2013 8:03 am

Cox regression - time variable

Postby csoki » Wed Aug 27, 2014 9:30 am


I can't find any information on this at all using my usual source (google), I hope someone can answer me! My statistics teacher only wrote in her notes that this is "relevant in any dataset", but she didn't write (or say) anything about how to solve the problem! Very useful... Thank you very much... :roll:

I'm going to do a survival analysis (Cox regression). I have dates for the following:
- start of follow-up
- event (disability pension)
- loss to follow-up (emigration)
- death
- retirement
- end of study (right censoring)
I need to combine these dates (or rather the days from start of follow-up to event or censoring) into one variable. But how do I do that? One subject can experience several of these events, but the event in question (disability pension) would be the first of them to occur, as you cannot get a disability pension if you are dead or have already retired. And if you emigrate, you won't end up in the disability pension registry.

I'm sure (or I hope) there is an easy way of doing this, but I simply can't find out what it is. :roll: Can someone help me please?
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Cox regression - time variable

Postby GerineL » Wed Aug 27, 2014 12:26 pm

I am not entirely sure what it is you want, but look into the date-time wizard, that might be of help!

Who is online

Users browsing this forum: No registered users and 1 guest