Surv {VGAM} | R Documentation |
Create a survival object, usually used as a response variable in a model formula.
Surv(time, time2, event, type =, origin = 0) is.SurvS4(x)
time |
for right censored data, this is the follow up time. For interval data, the first argument is the starting time for the interval. |
x |
any R object. |
event |
The status indicator, normally 0=alive, 1=dead. Other choices are
TRUE /FALSE (TRUE = death) or 1/2 (2=death). For
interval censored data, the status indicator is 0=right censored,
1=event at time , 2=left censored, 3=interval censored.
Although unusual, the event indicator can be omitted, in which case
all subjects are assumed to have an event.
|
time2 |
ending time of the interval for interval censored or counting
process data only. Intervals are assumed to be open on the left and
closed on the right, (start, end] . For counting process
data, event indicates whether an event occurred at the end of
the interval.
|
type |
character string specifying the type of censoring. Possible values
are "right" , "left" , "counting" ,
"interval" , or "interval2" . The default is
"right" or "counting" depending on whether the
time2 argument is absent or present, respectively.
|
origin |
for counting process data, the hazard function origin. This is most often used in conjunction with a model containing time dependent strata in order to align the subjects properly when they cross over from one strata to another. |
Typical usages are
Surv(time, event) Surv(time, time2, event, type=, origin=0)
In theory it is possible to represent interval censored data without a third column containing the explicit status. Exact, right censored, left censored and interval censored observation would be represented as intervals of (a,a), (a, infinity), (-infinity,b), and (a,b) respectively; each specifying the interval within which the event is known to have occurred.
If type = "interval2"
then the representation given above is
assumed, with NA taking the place of infinity. If `type="interval"
event
must be given.
If event
is 0
, 1
, or 2
, the relevant
information is assumed to be contained in time
,
the value in time2
is ignored, and the second column of the result will contain a
placeholder.
Presently, the only methods allowing interval censored data are the
parametric models computed by survreg
,
so the distinction between open and closed intervals
is unimportant.
The distinction is important for counting process data and
the Cox model.
The function tries to distinguish between the use of 0/1 and 1/2 coding for
left and right censored data using if (max(status)==2)
.
If 1/2 coding is used and all the subjects are censored, it will
guess wrong. Use 0/1 coding in this case.
An object of class SurvS4
(formerly Surv
).
There are methods for print
,
is.na
, and subscripting survival objects. SurvS4
objects
are implemented as a matrix of 2 or 3 columns.
In the case of is.SurvS4
, a logical value TRUE
if x
inherits from class "SurvS4"
, otherwise a FALSE
.
The purpose of having Surv
in VGAM is so that
the same input can be fed into vglm
as functions in
survival such as survreg
. The class
name has been changed from "Surv"
to "SurvS4"
; see
SurvS4-class
.
The format J+
is interpreted in VGAM as >= J.
If type="interval"
then these should not be used in VGAM:
(L,U-]
or (L,U+]
.
The code and documentation comes from survival.
Slight modifications have been made for conversion to S4 by T. W. Yee.
Also, for "interval"
data, as.character.SurvS4()
has
been modified to print intervals of the form
(start, end]
and not
[start, end]
as previously.
(This makes a difference for discrete data, such as for
cenpoisson
).
All VGAM family functions beginning with "cen"
require
the packaging function Surv
to format the input.
SurvS4-class
,
cenpoisson
,
survreg
,
leukemia
.
with(leukemia, Surv(time, status))