nodeapply: Apply functions over nodes

In etree: Classification and Regression with Structured and Mixed-Type Data

Apply functions over nodes

Description

Returns a list of values obtained by applying a function to "etree" or "partynode" objects.

Usage

nodeapply(obj, ids = 1, FUN = NULL, ...)

## S3 method for class 'partynode'
nodeapply(obj, ids = 1, FUN = NULL, ...)

## S3 method for class 'etree'
nodeapply(obj, ids = 1, FUN = NULL, by_node = TRUE, ...)

Arguments

obj	Object of class "etree" or "partynode".
ids	Integer vector of node identifiers to apply over.
FUN	Function to be applied to nodes. By default, the node itself is returned.
...	Additional arguments.
by_node	Logical indicating whether FUN should be applied to subsets of "partynode" objects (default) or not, in which case it is applied to subsets of "etree" objects.

Details

The method for "partynode" objects apply function FUN to all nodes with node identifiers in ids. The method for "etree" objects by default calls the nodeapply method on the corresponding node slot. If by_node is FALSE, it is applied to the "etree" object with root node ids.

Value

A list of results whose length is given by length(ids).

Methods (by class)

• partynode: nodeapply() method for objects of class "partynode".

• etree: nodeapply() method for objects of class "etree".

Examples

## Covariates
nobs <- 100
cov_num <- rnorm(nobs)
cov_nom <- factor(rbinom(nobs, size = 1, prob = 0.5))
cov_gph <- lapply(1:nobs, function(j) igraph::sample_gnp(100, 0.2))
cov_fun <- fda.usc::rproc2fdata(nobs, seq(0, 1, len = 100), sigma = 1)
cov_list <- list(cov_num, cov_nom, cov_gph, cov_fun)

## Response variable
resp_reg <- cov_num ^ 2

## Fit
etree_fit <- etree(response = resp_reg, covariates = cov_list)

## Get pvalues of inner nodes
tnodes <- nodeids(etree_fit, terminal = TRUE)
nodes <- 1:max(tnodes)
inodes <- nodes[-tnodes]
nodeapply(etree_fit, ids = inodes, FUN = function(n) n$info$pvalue)

etree documentation built on July 16, 2022, 9:05 a.m.