krylow.pls {timereg}R Documentation

Fits Krylow based PLS for additive hazards model

Description

Fits the PLS estimator for the additive risk model based on the least squares fitting criterion

L(β,D,d) = β^T D β - 2 β^T d

where D=int Z H Z dt and d=int Z H dN.

Usage

krylow.pls(D,d,dim)

Arguments

D defined above
d defined above
dim number of pls dimensions

Value

returns a list with the following arguments:

beta PLS regression coefficients

Author(s)

Thomas Scheike

References

Martinussen and Scheike, The Aalen additive hazards model with high-dimensional regressors, submitted.

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

Examples

## makes data for pbc complete case
data(pbc)
pbc$time<-pbc$time+runif(418); pbc$time<-pbc$time/365
pbc[pbc==-9]<-NA; rsum<-apply(pbc,1,sum); 
pbc<-pbc[!is.na(rsum),]
covs<-as.matrix(pbc[,-c(10,17)])
lcovs<-covs;
lcovs[,c(3,5,6,10,11,13,17,18)]<-
log(lcovs[,c(3,5,6,10,11,13,17,18)])

## computes the matrices needed for the least squares 
## criterion 
out<-aalen.test(Surv(time,status)~const(lcovs),pbc,robust=0,n.sim=0)
S=out$intZHZ; s=out$intZHdN;

out<-krylow.pls(S,s,2)

[Package timereg version 1.1-2 Index]