party (version 1.3-22)

plot.mob: Visualization of MOB Trees

Description

plot method for mob objects with extended facilities for plugging in panel functions.

Usage

# S3 method for mob
plot(x, terminal_panel = node_bivplot, tnex = NULL, ...)

Arguments

x

an object of class mob.

terminal_panel

a panel function or panel-generating function of class "grapcon_generator". See plot.BinaryTree for more details.

tnex

a numeric value giving the terminal node extension in relation to the inner nodes.

...

further arguments passed to plot.BinaryTree.

Details

This plot method for mob objects simply calls the plot.BinaryTree method, setting a different terminal_panel function by default (node_bivplot) and tnex value.

See Also

node_bivplot, node_scatterplot, plot.BinaryTree, mob

Examples

set.seed(290875)

if(require("mlbench")) {

## recursive partitioning of a linear regression model
## load data
data("BostonHousing", package = "mlbench")
## and transform variables appropriately (for a linear regression)
BostonHousing$lstat <- log(BostonHousing$lstat)
BostonHousing$rm <- BostonHousing$rm^2
## as well as partitioning variables (for fluctuation testing)
BostonHousing$chas <- factor(BostonHousing$chas, levels = 0:1, 
                             labels = c("no", "yes"))
BostonHousing$rad <- factor(BostonHousing$rad, ordered = TRUE)

## partition the linear regression model medv ~ lstat + rm
## with respect to all remaining variables:
fm <- mob(medv ~ lstat + rm | zn + indus + chas + nox + age + dis + 
                              rad + tax + crim + b + ptratio,
  control = mob_control(minsplit = 40), data = BostonHousing, 
  model = linearModel)

## visualize medv ~ lstat and medv ~ rm
plot(fm)

## visualize only one of the two regressors
plot(fm, tp_args = list(which = "lstat"), tnex = 2)
plot(fm, tp_args = list(which = 2), tnex = 2)

## omit fitted mean lines
plot(fm, tp_args = list(fitmean = FALSE))

## mixed numerical and categorical regressors 
fm2 <- mob(medv ~ lstat + rm + chas | zn + indus + nox + age + 
                                      dis + rad,
  control = mob_control(minsplit = 100), data = BostonHousing, 
  model = linearModel)
plot(fm2)

## recursive partitioning of a logistic regression model
data("SynthDiabetes", package = "mlbench")
fmSD <- mob(diabetes ~ glucose | pressure + triceps + insulin + mass + age,
  data = SynthDiabetes, model = glinearModel, 
  family = binomial())
## default plot: spinograms with breaks from five point summary
plot(fmSD)
## user-defined breaks (or: alternatively use breaks from hist() via fivenum = FALSE)
plot(fmSD, tp_args = list(breaks = c(0, 100, 125, 150, 175, 200)))
## CD plots instead of spinograms
plot(fmSD, tp_args = list(cdplot = TRUE))
## different smoothing bandwidth
plot(fmSD, tp_args = list(cdplot = TRUE, bw = 15))

}