relsurv (version 1.6-5)

epa: Excess hazard function smoothing

Description

An Epanechnikov kernel function based smoother for smoothing the baseline excess hazard calculated by the rsadd function with the EM method.

Usage

epa(fit,bwin,times,n.bwin=16,left=FALSE)

Arguments

fit

Fit from the additive relative survival model using the EM method.

bwin

The relative width of the smoothing window (default is 1).

times

The times at which the smoother is to be evaluated. If missing, it is evaluated at all event times.

n.bwin

Number of times that the window width may change.

left

If FALSE (default) smoothing is performed symmetrically, if TRUE only leftside neighbours are considered.

Value

• A list with two components:
• lambda
  the smoothed excess baseline hazard function
• times
  the times at which the smoothed excess baseline hazard is evaluated.

Details

The function performs Epanechnikov kernel smoothing. The follow up time is divided (according to percentiles of event times) into several intervals (number of intervals defined by n.bwin) in which the width is calculated as a factor of the maximum span between event times. Boundary effects are also taken into account on both sides.

References

Package. Pohar M., Stare J. (2006) "Relative survival analysis in R." Computer Methods and Programs in Biomedicine, 81: 272--278 Relative survival: Pohar, M., Stare, J. (2007) "Making relative survival analysis relatively easy." Computers in biology and medicine, 37: 1741--1749. EM algorithm: Pohar Perme M., Henderson R., Stare, J. (2009) "An approach to estimation in relative survival regression." Biostatistics, 10: 136--146.

See Also

rsadd,

Examples

data(slopop)
data(rdata)
#fit an additive model with the EM method
fit <- rsadd(Surv(time,cens)~sex+age+ratetable(age=age*365,
	   sex=sex,year=year), ratetable=slopop,data=rdata,int=5,method="EM")
sm <- epa(fit)
plot(sm$times,sm$lambda)