R: A direct interface to the 'computational engine' of...
survfitcoxph.fit
R Documentation
A direct interface to the ‘computational engine’ of survfit.coxph
Description
This program is mainly supplied to allow other packages to invoke the
survfit.coxph function at a ‘data’ level rather than a ‘user’ level.
It does no checks on the input data that is provided, which can lead
to unexpected errors if that data is wrong.
the response variable used in the Cox model. (Missing values
removed of course.)
x
covariate matrix used in the Cox model
wt
weight vector for the Cox model. If the model was unweighted
use a vector of 1s.
x2
matrix describing the hypothetical subjects for which a
curve is desired. Must have the same number of columns as x.
risk
the risk score exp(X beta) from the fitted Cox model. If
the model had an offset, include it in the argument to exp.
newrisk
risk scores for the hypothetical subjects
strata
strata variable used in the Cox model. This will be a factor.
se.fit
if TRUE the standard errors of the curve(s) are returned
survtype
1=Kalbfleish-Prentice, 2=Nelson-Aalen, 3=Efron. It is
usual to match this to the approximation for ties used in the
coxph model: KP for ‘exact’, N-A for ‘breslow’ and Efron for ‘efron’.
vartype
1=Greenwood, 2=Aalen, 3=Efron
varmat
the variance matrix of the coefficients
id
optional; if present and not NULL this should be
a vector of identifiers of length nrow(x2).
A mon-null value signifies that x2 contains time dependent
covariates, in which case this identifies which rows of x2 go
with each subject.
y2
survival times, for time dependent prediction. It gives
the time range (time1,time2] for each row of x2. Note: this
must be a Surv object and thus contains a status indicator, which is
never used in the routine, however.
strata2
vector of strata indicators for x2. This must
be a factor.
unlist
if FALSE the result will be a list with one
element for each strata. Otherwise the strata are “unpacked” into
the form found in a survfit object.
Value
a list containing nearly all the components of a survfit
object. All that is missing is to add the confidence intervals, the
type of the original model's response (as in a coxph object), and the
class.
Note
The source code for for both this function and
survfit.coxph is written using noweb. For complete
documentation see the inst/sourcecode.pdf file.