3

I am fairly new to survival analysis and am playing around in R. I have a fairly simple Cox model 

library(survival)
data(kidney)
cox<-coxph(Surv(time, type)~delta, data=kidney)
baseline <- basehaz(cox , centered=FALSE)
cox.survfit<- survfit(cox)
plot(cox.survfit)

enter image description here

My question is how do i calculate the survival rates myself without calling the survfit function. I tried to look at this stackoverflow post which i kind of understand but am not exactly able to turn it into code.

enter image description here

I have the h0 from basehaz but I am not sure about the rest of the calculations. Any ideas?

EDIT:- Added the kidney dataset in csv format(in case you don't have the same columns in your kidney datset ) :- 

"time","delta","type"
1.5,1,1
3.5,1,1
4.5,1,1
4.5,1,1
5.5,1,1
8.5,1,1
8.5,1,1
9.5,1,1
10.5,1,1
11.5,1,1
15.5,1,1
16.5,1,1
18.5,1,1
23.5,1,1
26.5,1,1
2.5,0,1
2.5,0,1
3.5,0,1
3.5,0,1
3.5,0,1
4.5,0,1
5.5,0,1
6.5,0,1
6.5,0,1
7.5,0,1
7.5,0,1
7.5,0,1
7.5,0,1
8.5,0,1
9.5,0,1
10.5,0,1
11.5,0,1
12.5,0,1
12.5,0,1
13.5,0,1
14.5,0,1
14.5,0,1
21.5,0,1
21.5,0,1
22.5,0,1
22.5,0,1
25.5,0,1
27.5,0,1
0.5,1,2
0.5,1,2
0.5,1,2
0.5,1,2
0.5,1,2
0.5,1,2
2.5,1,2
2.5,1,2
3.5,1,2
6.5,1,2
15.5,1,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
0.5,0,2
1.5,0,2
1.5,0,2
1.5,0,2
1.5,0,2
2.5,0,2
2.5,0,2
2.5,0,2
2.5,0,2
2.5,0,2
3.5,0,2
3.5,0,2
3.5,0,2
3.5,0,2
3.5,0,2
4.5,0,2
4.5,0,2
4.5,0,2
5.5,0,2
5.5,0,2
5.5,0,2
5.5,0,2
5.5,0,2
6.5,0,2
7.5,0,2
7.5,0,2
7.5,0,2
8.5,0,2
8.5,0,2
8.5,0,2
9.5,0,2
9.5,0,2
10.5,0,2
10.5,0,2
10.5,0,2
11.5,0,2
11.5,0,2
12.5,0,2
12.5,0,2
12.5,0,2
12.5,0,2
14.5,0,2
14.5,0,2
16.5,0,2
16.5,0,2
18.5,0,2
19.5,0,2
19.5,0,2
19.5,0,2
20.5,0,2
22.5,0,2
24.5,0,2
25.5,0,2
26.5,0,2
26.5,0,2
28.5,0,2
Nick Cox
  • 48,377
  • 8
  • 110
  • 156
user2912902
  • 31
  • 2

1 Answers1

1

basehaz does not actually give you $\hat{h}_0(t)$ but rather $\hat{H}_0(t)$, an estimate of the baseline cumulative hazard function. To get an estimate of the baseline hazard you need to take the differences or "jumps" in the cumulative hazard. Once you obtain this function you can then apply the standard Cox formula $\hat{h}_0(t) \exp(\beta^T x_i)$ to get an estimate of $\hat{h}_i(t)$.

dsaxton
  • 11,397
  • 1
  • 23
  • 45
  • you are right but i wanted to create the survival curve ....which i was able to with the following "plot(baseline$time , exp(-baseline$hazard * exp(testdata$delta * -0.2996697)))"......0.2996697 being the coeff of the column "delta" – user2912902 Jul 02 '15 at 22:06
  • Your expression is for the hazard function. The OP asked for the survival function. They are different. – AdamO Apr 06 '18 at 14:24