Statistics Toolbox™

nodeerr - Node errors

Class

@classregtree

Syntax

e = nodeerr(t) e = nodeerr(t,nodes)

Description

e = nodeerr(t) returns an n-element vector e of the errors of the nodes in the tree t, where n is the number of nodes. For a regression tree, the error e(i) for node i is the variance of the observations assigned to node i. For a classification tree, e(i) is the misclassification probability for node i.

e = nodeerr(t,nodes) takes a vector nodes of node numbers and returns the errors for the specified nodes.

The error e is the so-called resubstitution error computed by applying the tree to the same data used to create the tree. This error is likely to under estimate the error you would find if you applied the tree to new data. The test function provides options to compute the error (or cost) using cross-validation or a test sample.

Example

Create a classification tree for Fisher's iris data:

load fisheriris;

t = classregtree(meas,species,...
                 'names',{'SL' 'SW' 'PL' 'PW'})
t = 
Decision tree for classification
1  if PL<2.45 then node 2 else node 3
2  class = setosa
3  if PW<1.75 then node 4 else node 5
4  if PL<4.95 then node 6 else node 7
5  class = virginica
6  if PW<1.65 then node 8 else node 9
7  class = virginica
8  class = versicolor
9  class = virginica

view(t)

e = nodeerr(t)
e =
    0.6667
         0
    0.5000
    0.0926
    0.0217
    0.0208
    0.3333
         0
         0

Reference