coxphHaz {biostat3}

Smoothed hazard estimates for coxph

Description

Smoothed hazard estimates for coxph

Usage

coxphHaz(object, newdata, n.grid = 300, kernel = "epanechnikov", from,
to, ...)
## S3 method for class 'coxphHaz'
print(x, digits=NULL, ...)
## S3 method for class 'coxphHaz'
plot(x, xlab="Time", ylab="Hazard", type="l", ...)
## S3 method for class 'coxphHazList'
plot(x, xlab="Time", ylab="Hazard", type="l",
                  col=1:length(x), lty=1, legend.args=list(), ...)
## S3 method for class 'coxphHazList'
lines(x, ...)
## S3 method for class 'coxphHaz'
as.data.frame(x, row.names=NULL, optional=FALSE, level=0.95, ...)
## S3 method for class 'coxphHazList'
as.data.frame(x, row.names=NULL, optional=FALSE, ...)

Arguments

object

coxph object

newdata

data-frame with covariates for prediction

n.grid

the number of grid values for which the hazard is calculated

kernel

the kernel used for smoothing

from

argument for density. Defaults to the minimum time.

to

argument for density. Defaults to the maximum time.

x

object

digits

argument passed to print.density

col

graphics argument

lty

graphics argument

xlab

graphics argument

ylab

graphics argument

type

graphics argument

level

level for confidence intervals (default=0.95)

row.names

NULL or a character vector giving the row names for the data frame. Missing values are not allowed.

optional

logical. If TRUE, setting row names and converting column names (to syntactic names: see make.names) is optional. Note that all of R's base package as.data.frame() methods use optional only for column names treatment, basically with the meaning of data.frame(*, check.names = !optional). See also the make.names argument of the matrix method.

legend.args

a list of options that are passed to the legend call. Defaults are list(x="topright",legend=strata(attr(x,"newdata")),col=col,lty=lty).

...

other arguments. For coxphHaz, these arguments are passed to density. For the plot and lines methods, these are passed to the relevant plot, matplot and matlines functions.

Details

Smooth hazard estimates from a Cox model using kernel smoothing of the Nelson-Aalen estimator.

Value

The coxphHaz function returns either a class of type c("coxphHaz","density") when newdata has one row or, for multiple rows in newdata, a class of type "coxphHazList", which is a list of type c("coxphHaz","density").

See Also

coxph, survfit, density

Examples

fit <- coxph(Surv(surv_mm/12,status=="Dead: cancer")~agegrp, data=colon)
newdata <- data.frame(agegrp=levels(colon$agegrp))
haz <- suppressWarnings(coxphHaz(fit,newdata))
plot(haz, xlab="Time since diagnosis (years)")