XploRe Help : grcarttree

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library:	xclust
See also:	cartsplit dispcarttree grcarttree grcart2

Macro:		dispcarttree
Description:		Plots the regression tree of CART in a display

Reference(s):

Breiman, L., Friedman, J., Olshen, R., and Stone, C. J. (1984). Classification and Regression Trees. Chapman & Hall, New York

Link:

CART Tutorial

Usage:	`{g, gp} = grcarttree(ts,stri)`
Input:
	`cs`	list of vectors: data structure which represents a binary tree and is produced by cartsplit procedure, contains vectors .. cs.val, cs.vec, cs.mean, cs.ssr, cs.nelem. See cartsplit for the description of cs.
	`stri`	string: optional argument, either missing, "nelem", "ssr" or "mean".
Output:
	`g`	CART tree
	`gp`	text labels for CART tree

Note:

The content of the nodes in the picture of the tree is determined by the parameter stri. If this parameter is not given, in the nonterminal nodes there are given the splitting rule (for example x1<=1.34), and in the leaf nodes there are given the values of the regression function, that is, the mean of the observations in that node. If stri = "nelem", the number of observations in the nodes is plotted, if stri = "ssr", the sum of squared residuals in the nodes is plotted, and if stri = "mean", the mean of observations in the nodes is plotted.

Example:

; loads the library xclust library ("xclust") ; generate an artifcal dataset x1=#(0,0,0,0,1,1,1,1,1,2) x2=#(0,0,0,0,0,0,0,1,1,1) x=x1~x2 y=#(0,0,0,0,100,100,100,120,120,120) ; run CART on our data cs=cartsplit(x,y,#(0,1)) ; plot CART tree d = createdisplay(1,1) {g, gp} = grcarttree (cs) show (d, 1, 1, g, gp)

Result:

shows the tree with the splitting rules

Library:	xclust
See also:	cartsplit dispcarttree grcarttree grcart2

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Author: 1999 Jussi Klemelae, 2000-04-15, Sigbert Klinke