Survival Models - Cox PH

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Survival Models - Cox PH

Hema Dave
Dear List,

I am working on Survival model (cox ph) and I am stuck with the model
prediction. Need guidance.

If i understand correctly, for the cox ph models, SPSS predicts the hazard
rate for agiven time. My questions are
1. How can I get the predicted probabilities of an event.
2. Secondly, i want gain charts, similar to the ones i get in logistic
regression, how can i get that? I mean, what should the deciling be on?
Since the hazard rate that i get from SPSS, increases as the time
increases, hence if i do the deciling on hazard rate i will never get a
lift kind of graph.
3. Is there a way to calculate the time factor usingthe predicted hazard
rate?

Help very much appreciated!
Regards,
Hema

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD