wfu: Weight function (for Mahalabobis distances)

In fpc: Flexible Procedures for Clustering

Weight function (for Mahalabobis distances)

Description

Function of the elements of md, which is 1 for arguments smaller than ca, 0 for arguments larger than ca2 and linear (default: continuous) in between.

Thought for use in fixmahal.

Usage

wfu(md, ca, ca2, a1 = 1/(ca - ca2), a0 = -a1 * ca2)

Arguments

md	vector of positive numericals.
ca	positive numerical.
ca2	positive numerical.
a1	numerical. Slope.
a0	numerical. Intercept.

Value

A vector of numericals between 0 and 1.

Author(s)

Christian Hennig [e-mail] https://www.unibo.it/sitoweb/christian.hennig/en/

See Also

fixmahal

Examples

  md <- seq(0,10,by=0.1)
  round(wfu(md,ca=5,ca2=8),digits=2)

fpc documentation built on Sept. 24, 2024, 9:07 a.m.