Learn R Programming

mvtboost (version 0.5.0)

mvtb.perspec: Perspective plot for 2 predictors and 1 response.

Description

This is a plot of the model implied function of 2 predictors averaged over the other predictors included in the model. This is called a partial dependence plot. As an alternative to the perspective (3D) plot, a 2D heat plot can be obtained directly using ?plot.gbm.

Usage

mvtb.perspec(object, response.no = 1, predictor.no = 1:2, n.trees = NULL, phi = 15, theta = -55, r = sqrt(10), d = 3, xlab = NULL, ylab = NULL, zlab = NULL, ticktype = "detailed", ...)

Arguments

object
mvtb output object
response.no
index of the response variable
predictor.no
vector containing indices of the predictor variables to plot
n.trees
desired number of trees. Defaults to the minimum number of trees by CV, test, or training error
phi
angle of viewing direction. See ?persp.
theta
angle of viewing direction See ?persp.
r
distance from eye to center. See ?persp.
d
strength of perspective. See ?persp.
xlab,
title for x axis, must be character strings.
ylab,
title for y axis, must be character strings.
zlab,
title for z axis, must be character strings.
ticktype
'detailed' gives axis points. See ?persp for other options.
...
extra arguments are passed to persp. See ?persp

Value

Function returns a plot.

See Also

plot.gbm, plot.mvtb, mvtb.heat