balance (version 0.2.4)

sbp.fromPDBA: Build SBP Matrix of "Principal Discriminant Balances"

Description

This function builds an SBP of "discriminant balances" by clustering a matrix of the pair-wise total within-group variance, adjusted by the pair-wise total variance (inverted by subtracting this value from 1). The method is intended to make the largest balances most discriminative.

Usage

sbp.fromPDBA(x, group, ...)

Arguments

x

A matrix with rows as samples (N) and columns as components (D).

group

A character vector. Group or sub-group membership. Argument passed to propr::propd.

...

Other arguments passed to propr::propd.

Value

An SBP matrix.

Examples

# NOT RUN {
library(balance)
data(iris)
x <- iris[1:100,1:4]
y <- iris[1:100,5]
sbp.fromPDBA(x, y)
# }
# NOT RUN {
# }