ecif.coef: Excess CIF prediction based on newdata and geepack::geese...

In cribosch/matchsurv: Excess Risk Estimation for Matched Cohort Survival Data

Description

Excess CIF prediction based on newdata and geepack::geese estimates.

Usage

ecif.coef(model, times, link = "log", level = 0.95)

Arguments

model	geese object. To be defined if coefs and vcov are null.
times	vector of timepoints as the one used to estimate the GEE model
link	link used to estimate the model. Choose between c("log","logit","identity")
level	confindence interval level, default=0.95

Value

table with coefficient estimates, sandwich standard error, function of coefficient for interpretation and p-value.

Examples

dcif<-sim.data.MatchCR(1000,5)
tp<-c(0.5,1,2,5,10,15,25)
setdcif1<-compcomp(timereg::Event(time=FALSE,time2=time,cause=cause)~X1+X2, 
                   data=dcif, cluster=i, idControl=j, time.points=tp,
                   cens.formula=NULL, event=1)
exc.cif.mod1<-geepack::geese(Rt~-1+factor(h)+X1+X2,
                             data=setdcif1,
                             family="gaussian", #error distribution
                             mean.link = "log", #link function for Rt
                             corstr="independence", #correlation structure
                             id=clust.num, #cluster vector
                             weights=weights #censoring weights 
                             )
ecif.coef(exc.cif.mod1,times = tp, link = "log")

cribosch/matchsurv documentation built on Aug. 15, 2019, 11:55 a.m.