censorReg
is an alternative function for fitting this type
of model.
survreg
function is deprecated; please use
instead.
survreg(formula=formula(data), data=sys.parent(), subset, na.action, link="log", dist="extreme", init=NULL, fixed=list(), control, model=F, x=F, y=T, ...)
Surv
function.
See the documentation for
Surv
,
lm
and
formula
for details.
formula
, or the
subset
argument.
subset
argument has been used. Default is
options()$na.action
.
"log"
and
"identity"
.
"extreme"
,
"logistic"
,
"gaussian"
,
"exponential"
,
and
"rayleigh"
.
maxiter
,
rel.tolerance
and
failure
)
returned by
survreg.control
.
See
survreg.control
and the EXAMPLES section for details.
TRUE
then the model frame is included as component
model
in the object returned by the function.
TRUE
then the model matrix is returned as component
x
in the object returned by the function.
TRUE
then the response is returned as component
y
in the object returned by the function.
lm
, including
singular.ok
.
survreg
is returned, which inherits from class
glm
.
See
survreg.object
for details.
This routine is not as robust against nearly singular X matrices as
lm
; the problem occurs when the covariance matrix is explicitly inverted
using the
solve
function. This can sometimes be solved by subtracting
the mean from all continuous covariates.
# Fit an exponential model survreg(Surv(futime, fustat) ~ ecog.ps + rx, data=ovarian, dist='extreme', link='log', fixed=list(scale=1)) # Using survreg.control to supply the control argument survreg(Surv(days, event) ~ voltage, data=capacitor, control=survreg.control(maxiter = 40))