Pass in a tree and a node number. This returns a vector of booleans identifying which members of the training set
belong to the given region.
Mainly used to form vectors nu to define parameters from a given tree. Since this is called internally by branchInference()
,
will rarely be needed directly by users.
getRegion(tree, nn)
tree | An rpart object. Must have been built with model=TRUE |
---|---|
nn | A node number. Can be a string or an integer. |
The indices of data that belong to this region in the training set. The training set is stored in tree$model.
data(blsdata, package="treevalues") bls.tree <-rpart::rpart(kcal24h0~hunger+disinhibition+resteating+rrvfood+liking+wanting, model = TRUE, data = blsdata, cp=0.02) branch <- getBranch(bls.tree, 2) left_child <- getRegion(bls.tree,2) right_child <- getRegion(bls.tree,3) nu_sib <- left_child/sum(left_child) - right_child/sum(right_child) S_sib <- getInterval(bls.tree, nu_sib,branch) correctPVal(S_sib, nu_sib, blsdata$kcal24h0, sd(blsdata$kcal24h0))#> [1] 0.4425364