boundary {exactmaxsel} | R Documentation |
The function boundary
is an internal function that computes the greatest (and also the smallest if
lower=TRUE
) number of observations
from class Y=1 in the left node that lead to an association criterion <= c.
Remark: in general, the numbers output by boundary
are not integers.
boundary(x, n0, n1, c, statistic, lower=TRUE)
x |
the number of observations in the left node |
n0 |
the number of observations in class Y=0 |
n1 |
the number of observations in class Y=1 |
c |
the value of the criterion that should not be exceeded. |
statistic |
the association measure. Currently, only
statistic="chi2" (chi-square statistic)
and statistic="gini" (the Gini-gain from machine learning) are
implemented. |
lower |
Should the lower boundary also be computed? |
This function should not be called by the user in practice.
a list with
upper |
the upper boundary (greatest allowed value). |
lower |
the lower boundary (smallest allowed value). |
Anne-Laure Boulesteix (http://www.slcmsr.net/boulesteix).
A.-L. Boulesteix (2006), Maximally selected chi-square statistics for ordinal variables, Biometrical Journal 48:451-462.
A.-L. Boulesteix (2006), Maximally selected chi-square statistics and binary splits of nominal variables, Biometrical Journal 48:838-848.
C. Strobl, A.-L. Boulesteix and T. Augustin (2007), Unbiased split selection for classification trees based on the Gini index, Computational Statistics and Data Analysis (in press).
# load exactmaxsel library library(exactmaxsel) boundary(10,30,30,c=3,statistic="chi2",lower=TRUE)