Usage: |
{data,ties} = hazdat(t, delta {,z})
|
Input: |
| t | n x 1 vector, the censored survival time.
|
| delta | n x 1 vector, the censoring indicator, showing
if censoring occured (0) or not (1).
|
| z | optional, n x p matrix of covariates (default is an empty matrix):
each row denotes the p-dimensional
covariate vector associated with the i-th individual
|
Output: |
| data | n x (p+4) matrix of cosorted time-to-event data:
column 1: sorted observed times t_i
column 2: cosorted censoring indicator delta_i
column 3: cosorted original observation labels i (i=1,...,n)
column 4: cosorted number of tied observations at time t_i
columns 5 through (p+4): cosorted covariate matrix z |
| ties | scalar, either 1 (there are ties) or 0 (there are no ties) |
- Example:
library("hazreg")
randomize(1)
y = -log(1-uniform(20)) ; exponential survival
c = 2*uniform(20) ; uniform censoring
t = min(y~c,2) ; censored time
delta =(y<=c) ; censoring indicator
res= hazdat(t,delta) ; preparing data
res
- Result:
The censored data is sorted and information about
ties in the data presented:
Contents of res.data
[ 1,] 0.0004326 1 1 1
[ 2,] 0.010333 0 13 1
[ 3,] 0.024494 1 3 1
[ 4,] 0.036793 1 5 1
.
.
.
[18,] 0.72464 0 2 1
[19,] 1.3491 1 7 1
[20,] 1.5549 0 20 1
Contents of res.ties
[1,] 0
- Example:
library("hazreg")
y = 2|1|3|2|4|7|1|3|2 ; hypothetical survival
c = 3|1|5|6|1|6|2|4|5 ; hypothetical censoring
t = min(y~c,2) ; censored time
delta =(y<=c) ; censoring indicator
res = hazdat(t,delta) ; preparing data
res
- Result:
The censored data is sorted and this time there are
ties in the data: three 1's, three 2's, two 3's:
Contents of res.data
[1,] 1 0 5 3
[2,] 1 1 7 3
[3,] 1 1 2 3
[4,] 2 1 4 3
[5,] 2 1 9 3
[6,] 2 1 1 3
[7,] 3 1 8 2
[8,] 3 1 3 2
[9,] 6 0 6 1
Contents of res.ties
[1,] 1