comparetree: comparetree

In paytonjjones/networktree: Recursive Partitioning of Network Models

comparetree

Description

Quickly compares two partitions of a networktree object

Usage

comparetree(
  tree,
  id1 = 2L,
  id2 = 3L,
  transform = "detect",
  highlights = 5,
  plot = FALSE,
  plot.type = c("compare", "subtract"),
  layout = "constrained",
  ...
)

Arguments

tree	a networktree object
id1	the first partition
id2	the second partition
transform	should stored correlation matrices be transformed to partial correlations or graphical lasso? Can be set to "cor", "pcor", or "glasso". Defaults to automatic detection
highlights	the number of comparisons to highlight
plot	plot a comparison of the two partitions?
plot.type	"compare" or "subtract". "compare" plots the two networks side by side. "subtract" subtracts network 2 from network 1, and plots a network where edge weights indicate the difference
layout	layout for the plots. The default "constrained" uses a FR layout from the full dataset
...	additional arguments passed to qgraph

Examples

set.seed(1)
d <- data.frame(trend = 1:200, foo = runif(200, -1, 1))
d <- cbind(d, rbind(
  mvtnorm::rmvnorm(100, mean = c(0, 0, 0),
          sigma = matrix(c(1, 0.5, 0.5, 0.5, 1, 0.5, 0.5, 0.5, 1), ncol = 3)),
  mvtnorm::rmvnorm(100, mean = c(0, 0, 0),
          sigma = matrix(c(1, 0, 0.5, 0, 1, 0.5, 0.5, 0.5, 1), ncol = 3))
))
colnames(d)[3:5] <- paste0("y", 1:3)

## Generate a networktree
tree1 <- networktree(nodevars=d[,3:5], splitvars=d[,1:2])

## Print out the structure
tree1

## Compare any two partitions
comparetree(tree1, id1=2, id2=3, highlights=3)

## Add a comparison plot
comparetree(tree1, id1=2, id2=3, plot=TRUE)

paytonjjones/networktree documentation built on Sept. 14, 2022, 8:01 a.m.