[R] Estimating Daily Survival
Bret Collier
bret at tamu.edu
Mon Mar 20 14:22:50 CET 2006
R Users,
I was wondering if someone might point me in the right direction. I am using a Cox model (survival package) to evaluate survival of pen-reared birds (time to event data collected daily) and I have been trying to determine how I can estimate a 'daily' survival rate and std error from the results of a Cox model?
Using survfit (see input data below) I computed the predicted survivor function for a Cox model, but I have been unable to figure out how to estimate a 'daily' survival rate for an average individual? Should I be looking at survexp? Any suggestions would be appreciated.
Thanks,
Bret
brettest1<-coxph(Surv(Entry, Exit, Fate)~Sex , data=mydat)
brettest1
coef exp(coef) se(coef) z p
Sex -0.503 0.605 0.524 -0.96 0.34
Likelihood ratio test=0.92 on 1 df, p=0.337 n= 19
> survfit(brettest1)$surv
[1] 0.9502716 0.9010356 0.8508737 0.7996929 0.7473818 0.6957615 0.5872297 0.5324263 0.4756545
[10] 0.4201535 0.3547775 0.2916481 0.2189030 0.1374015
>mydat
ID Year Dayrelease Agerelease Survivorship Entry Exit Fate Sex
1 16240 1996 205 95 164 205 369 1 0
2 16319 1996 205 88 140 205 345 1 0
3 16378 1996 248 108 100 248 348 1 0
4 20383 1996 241 98 204 241 445 1 0
5 16324 1996 219 90 227 219 446 1 0
6 16327 1996 219 90 497 219 716 1 0
7 20373 1996 241 114 413 241 654 1 0
8 20374 1996 241 111 211 241 452 1 0
9 16241 1996 205 95 234 205 439 1 1
10 16321 1996 219 90 118 219 337 1 1
11 16323 1996 219 90 180 219 399 1 1
12 20375 1996 241 103 268 241 509 1 1
13 20384 1996 241 98 299 241 540 1 1
14 20390 1996 241 93 204 241 445 1 1
15 20393 1996 241 88 208 241 449 1 1
16 16313 1996 248 122 512 248 760 0 1
17 20378 1996 241 103 236 241 477 0 1
18 20381 1996 241 101 329 241 570 0 1
19 16328 1996 219 90 224 219 443 0 1
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major 2
minor 2.1
year 2005
month 12
day 20
svn rev 36812
language R
More information about the R-help
mailing list