Tree structure obtained from combining mixture components

The method implemented in clustCombi can be used for combining Gaussian mixture components for clustering. This provides a hierarchical structure which can be graphically represented as a tree.

Usage

combiTree(object, type = c("triangle", "rectangle"),
                  yaxis = c("entropy", "step"), 
                  edgePar = list(col = "darkgray", lwd = 2), 
                  ...)

Arguments

object: An object of class 'clustCombi' resulting from a call to clustCombi.
type: A string specifying the dendrogram's type. Possible values are "triangle" (default), and "rectangle".
yaxis: A string specifying the quantity used to draw the vertical axis. Possible values are "entropy" (default), and "step".
edgePar: A list of plotting parameters. See dendrogram.
...: Further arguments passed to or from other methods.

Value

The function always draw a tree and invisibly returns an object of class 'dendrogram' for fine tuning.

Author

L. Scrucca

Examples

# \donttest{
data(Baudry_etal_2010_JCGS_examples)
output <- clustCombi(data = ex4.1) 
combiTree(output)

combiTree(output, type = "rectangle")

combiTree(output, yaxis = "step")

combiTree(output, type = "rectangle", yaxis = "step")

# }