rug.tree(tree, var, FUN = mean, ..., screen.arg = <<see below>>)
tree.
This is assumed to be the result of some function that produces
an object with the same named
components as that returned by the
tree function.
tree was made, or any numeric
vector of the same length.
var at each terminal node.
FUN.
split.screen and
tree.screens for more details.
FUN(var) for the
observations in each leaf.
FUN(var) lined up under each leaf of the dendrogram.
Assumes that
tree.screens has been used to divide the graphics window.
This is currently NOT a method of the
rug function.
tree.screens() z.auto <- tree(Mileage ~ Weight, car.test.frame) plot(z.auto) rug.tree(z.auto, HP) #heights show average HP for cars in each leaf node zo <- order.tree(z.auto, Mileage) #reorder leaf nodes based on Mileage screen(1) #clear screen 1 for new plot plot(zo) text(zo, pretty = T) rug.tree(zo, Price) #graphically assess correlation of Mileage and Price across leaves