cccd (version 1.00.05)

ccd: Cluster Catch Digraphs

Description

construct the cluster catch digraph from a data matrix.

Usage

ccd(data, m = 1, alpha = 0.05, sequential = TRUE)

Arguments

data

a matrix of observations

m

slope of the null hypothesis curve

alpha

alpha for the K-S test if sequential=T

sequential

use the sequential or non-sequential version

Value

• an object of class igraph. In addition, this contains the attributes:
• R
  the radii.
• stats
  the K-S statistics.
• layout
  the data vectors.
• walks
  the y-values of the random walks.
• fs
  the null hypothesis curve.
• A
  the adjacency matrix.
• m,alpha
  arguments passed to ccd.

Details

cluster cover digraph.

References

D.J. Marchette, Random Graphs for Statistical Pattern Recognition, John Wiley & Sons, 2004.

See Also

cccd

Examples

x <- matrix(rnorm(100),ncol=2)
G <- ccd(x)
plotCCD(G)