Recursive Partitioning and Regression Trees

Wrapper function for fitting and plotting rpart models

Usage

mvpart(form, data, minauto = TRUE, size, xv = c("1se", "min",
    "pick", "none"), xval = 10, xvmult = 0, xvse = 1, snip = FALSE,
    plot.add = TRUE, text.add = TRUE, digits = 3, margin = 0,
    uniform = FALSE, which = 4, pretty = TRUE, use.n = TRUE,
    all.leaves = FALSE, bars = TRUE, legend, bord = FALSE,
    xadj = 1, yadj = 1, prn = FALSE, branch = 1, rsq = FALSE,
    big.pts = FALSE, pca = FALSE, interact.pca = FALSE,
    wgt.ave.pca = FALSE, keep.y = TRUE, ...)

Arguments

form

As for rpart function. Arguments to rpart can be passed by ....

data

Optional data frame in which to interpret the variables named in the formula

minauto

If TRUE uses smart minsplit and minbucket based on N cases.

size

The size of tree to be generated.

xv

Selection of tree by cross-validation: "1se" - gives best tree within one SE of the overall best, "min" - the best tree, "pick" - pick the tree size interactively, "none" - no cross-validation.

xval

Number of cross-validations or vector defining cross-validation groups.

xvmult

Number of multiple cross-validations.

xvse

Multiplier for the number of SEs used for xv = "1se".

plot.add

Plot the tree and (optionally) add text.

text.add

Add output of text.rpart to tree.

snip

Interactively prune the tree.

digits

Number of digits on labels.

margin

Margin around plot, 0.1 gives an extra 10 percent space around the plot.

uniform

Uniform lengths to the branches of the tree.

which

Which split labels and where to plot them, 1=centered, 2 = left, 3 = right and 4 = both.

pretty

Pretty labels or full labels.

use.n

Add number of cases at each node.

all.leaves

Annotate all nodes.

bars

If TRUE adds barplots to nodes.

legend

If TRUE adds legend for mrt and classification trees.

bord

Border (box) around the barplots.

xadj, yadj

Adjust the size of the individual barplots (default = 1).

prn

If TRUE prints tree details.

branch

Controls spread of branches: 1=vertical lines, 0=maximum slope.

rsq

If TRUE gives "rsq" plot.

big.pts

Plot colored points at leaves – useful to link to PCA plot.

pca

If TRUE plots PCA of group means and add species and site information.

interact.pca

If TRUE runs interactive PCA. See rpart.pca.

wgt.ave.pca

If TRUE plot weighted averages acorss sites for species.

keep.y

If TRUE y values are returned.

...

... other arguments passed to rpart.

Value

an object of class rpart, a superset of class tree.

See also

rpart, rpart.pca,

Examples

data(spider)
mvpart(data.matrix(spider[,1:12])~herbs+reft+moss+sand+twigs+
water,spider)       # defaults

mvpart(data.matrix(spider[,1:12])~herbs+reft+moss+sand+twigs+
water,spider,xv="p")  # pick the tree size


# pick cv size and do PCA
fit <- mvpart(data.matrix(spider[,1:12])~herbs+reft+moss+sand+
twigs+water,spider,xv="1se",pca=TRUE)


rpart.pca(fit,interact=TRUE,wgt.ave=TRUE)

# interactive PCA plot of saved multivariate tree