timereg (version 1.8.6)

pe.sasieni: Fits Proportional excess hazards model with fixed offsets

Description

Fits proportional excess hazards model. The Sasieni proportional excess risk model.

The models are written using the survival modelling given in the survival package.

Usage

pe.sasieni(formula=formula(data),data=sys.parent(),
id=NULL,start.time=0,max.time=NULL,offsets=0,Nit=50,detail=0,n.sim=500)

Arguments

formula

a formula object, with the response on the left of a `~' operator, and the terms on the right. The response must be a survival object as returned by the `Surv' function.

data

a data.frame with the variables.

id

gives the number of individuals.

start.time

starting time for considered time-period.

max.time

stopping considered time-period if different from 0. Estimates thus computed from [0,max.time] if max.time>0. Default is max of data.

offsets

fixed offsets giving the mortality.

Nit

number of itterations.

detail

if detail is one, prints iteration details.

n.sim

number of simulations, 0 for no simulations.

Value

• Returns an object of type "pe.sasieni". With the following arguments:
• cum
  baseline of Cox model excess risk.
• var.cum
  pointwise variance estimates for estimated cumulatives.
• gamma
  estimate of relative risk terms of model.
• var.gamma
  variance estimates for gamma.
• Ut
  score process for Cox part of model.
• D2linv
  The inverse of the second derivative.
• score
  final score
• test.Prop
  re-sampled absolute supremum values.
• pval.Prop
  p-value based on resampling.

Details

The program assumes that there are no ties, and if such are present random noise is added to break the ties.

References

Martinussen and Scheike, Dynamic Regression Models for Survival Data, Springer Verlag (2006).

Sasieni, P.D., Proportional excess hazards, Biometrika (1996), 127--41.

Cortese, G. and Scheike, T.H., Dynamic regression hazards models for relative survival (2007), submitted.

Examples

data(mela.pop)
out<-pe.sasieni(Surv(start,stop,status==1)~age+sex,mela.pop,
id=1:205,Nit=10,max.time=7,offsets=mela.pop$rate,detail=0,n.sim=100)
summary(out)

ul<-out$cum[,2]+1.96*out$var.cum[,2]^.5
ll<-out$cum[,2]-1.96*out$var.cum[,2]^.5
plot(out$cum,type="s",ylim=range(ul,ll))
lines(out$cum[,1],ul,type="s"); lines(out$cum[,1],ll,type="s")
# see also prop.excess function