04-28-2017 05:01 AM
I try to run a shared frailty model with this code :
proc phreg data=ptiteq_c.c2 ; class cut equipgrf; model delaij*evtgrf(0)= cut /rl; random equipgrf/ solution ; run;
What I get is this :
Apparently, I have to take into account adjusted p-values from type 3 tests instead of usual p-values ("Penalized Partial Likelihood Approach "). I am also interested in the value of hazard ratio and 95% CI of my parameter "cut" (fixed effect). However, HR and 95%CI correspond to a usual p-value and not an adjusted p-value. As a result, I don't know if HR and 95%CI estimations are correct or not and if I can present adjusted p-values, HR and 95%CI together in the same table.
Could you help me please?