uni.logrank {uni.survival.tree} | R Documentation |
Univariate binary splits by the logrank test
Description
The output is the summary of significance tests for binary splits, where the cut-off values are optimized for each covariate.
Usage
uni.logrank(t.vec, d.vec, X.mat)
Arguments
t.vec |
:Vector of survival times (time to either death or censoring) |
d.vec |
:Vector of censoring indicators (1=death, 0=censoring) |
X.mat |
:n by p matrix of covariates, where n is the sample size and p is the number of covariates |
Details
The output can be used to construct a logrank tree.
Value
A dataframe containing:
Pvalue: the P-value of the two-sample logrank test, where the cut-off value is optimized
cut_off_point: the optimal cutt-off values of the binary splits given a feature
left.sample.size: the sample size of a left child node
right.sample.size: the sample size of a right child node
Examples
data(Lung,package="compound.Cox")
train_Lung=Lung[which(Lung[,"train"]==TRUE),] #select training data
t.vec=train_Lung[,1]
d.vec=train_Lung[,2]
x.mat=train_Lung[,-c(1,2,3)]
uni.logrank(t.vec,d.vec,x.mat)
[Package uni.survival.tree version 1.5 Index]