getTree.icrf {icrf}

Extract a single tree from an icrf object

Description

getTree 'extracts the structure of a tree from an icrf object.' Among nfold forests, the forest designated when implementing icrf will be considered. i.e., the kth tree of the last forest, when returnBest = FALSE or the tree of the best forest, when returnBest = TRUE, will be extracted. (Quoted statements are from randomForest by Liaw and Wiener unless otherwise mentioned.)

Usage

getTree(x, ...)

## S3 method for class 'icrf'
getTree(x, k = 1, labelVar = FALSE, ...)

Arguments

x

an icrf object.

...

not used.

k

'which tree to extract?'

labelVar

Splitting variables will be labelled with the original names when labelVar = TRUE. Otherwise they will be expressed as integers.

Details

'For numerical predictors, data with values of the variable less than or equal to the splitting point go to the left daughter node.'

'For categorical predictors, the splitting point is represented by an integer, whose binary expansion gives the identities of the categories that goes to left or right. For example, if a predictor has four categories, and the split point is 13. The binary expansion of 13 is (1, 0, 1, 1) (because 13 = 1*2^0 + 0*2^1 + 1*2^2 + 1*2^3), so cases with categories 1, 3, or 4 in this predictor get sent to the left, and the rest to the right.'

Value

'A matrix (or data frame, if labelVar=TRUE) with' (5 + number of time points) 'columns and number of rows equal to total number of nodes in the tree. The columns are:'

• left daughter: 'the row where the left daughter node is; 0 if the node is terminal'

• right daughter: 'the row where the right daughter node is; 0 if the node is terminal'

• split var: 'which variable was used to split the node; 0 if the node is terminal'

• split point: 'where the best split is; see Details for categorical predictor'

• status: 'is the node terminal' (-1) or not (-3)

• from the 6th to the last columns: the survival probability prediction for the node; each column represents the distinct time points. '0 if the node is not terminal'

Examples

library(survival) # for Surv()
data(rat2)
L = ifelse(rat2$tumor, 0, rat2$survtime)
R = ifelse(rat2$tumor, rat2$survtime, Inf)
# Note that this is a toy example. Use a larger ntree and nfold in practice.

set.seed(1)
rats.icrf <-
  icrf(Surv(L, R, type = "interval2") ~ dose.lvl + weight + male + cage.no,
       data = rat2, ntree = 10, nfold = 3)
getTree(rats.icrf, k = 2)