uniCoxCV {uniCox}R Documentation

Function to cross-validate a high dimensional Cox survival model using Univariate Shrinkage

Description

Function to cross-validate a high dimensional Cox survival model using Univariate Shrinkage

Usage

uniCoxCV(fit,x,y,status,nfolds=5,folds=NULL)

Arguments

fit object returned by call to uniCox
x Feature matrix, n obs by p variables
y Vector of n survival times
status Vector of n censoring indicators (1= died or event occurred, 0=survived, or event was censored)
nfolds Number of cross-valdiation folds
folds Optional list of sample numbers defining folds

Details

This function does cross-validation for a prediction model for survival data with high-dimensional covariates, using the Unvariate Shringae method.

Value

A list with components

devcvm Average drop in CV deviance for each lambda value
ncallcvm=ncallcvm Average number of features with non-zero wts in the CV, for each lambda value
se.devcvm Standard error of average drop in CV deviance for each lambda value
devcv Drop in CV deviance for each lambda value
ncallcv Number of features with non-zero wts in the CV, for each lambda value
folds Indices for CV folds
call Call to this function

Source

Tibshirani, R. Univariate shrinkage in the Cox model for high dimensional data (2009). http://www-stat.stanford.edu/~tibs/ftp/cus.pdf To appear SAGMB.

Examples

library(survival)
# generate some data
x=matrix(rnorm(200*1000),ncol=1000)
y=abs(rnorm(200))
x[y>median(y),1:50]=x[y>median(y),1:50]+3
status=sample(c(0,1),size=200,replace=TRUE)

# fit uniCox model
a=uniCox(x,y,status)

# do cross-validation to examine choice of lambda
aa=uniCoxCV(a,x,y,status)

[Package uniCox version 1.0 Index]