Time dependent cox model in PHREG

[Sathish_jammy]

Hi Experts,

I need to do an analysis of the Time-dependent cox model for the given dataset.

I get an error in the PROC PHREG procedure for the given dataset model.

Kindly suggest to me how to deal with the time-dependent cox model with the proper code to get a result.

77 proc phreg data=cox;
78 model End_Dur*EPI_Log(1)
79 run;
ERROR 73-322: Expecting an =.

proc phreg data=cox;
model End_Dur*EPI_Log(1)
run;

I considered the EPI_stage (1,2) as Log '0'(Normal) and EPI_stage (3) as Log '1'(Event) 

End_Dur in Years

[SteveDenham · Accepted solution]

Well, for a MODEL statement, the syntax is:

• MODEL response <*censor (list)> = effects </ options>;

Or

• MODEL (t1, t2)<*censor(list)> = effects </ options>;

So you need some effect variables that you wish to test for their influence on the response variable..  Looking through the data shown, I don't see any variables that would seem to fall into this category, so I don't know what to suggest.

SteveDenham