modelKS-class: 'modelKS' is a class of parameters of (K, S) model.

In ClustMMDD: Variable Selection in Clustering by Mixture Models for Discrete Data

Description

modelKS is a class that can contain the set of parameters associated to a model given by (K, S).

Objects from this class

Objects can be created by calling new("modelKS", ...). See new for more details.

Slots

N:

The number of individuals in the datatset.

P:

The number of random variables considered in the dataset.

N_levels:

A vector of the numbers of levels for the considered variables.

levels:

A "list" of the observed levels for the variables.

K:

The number of clusters.

S:

A vector of "logical" indicating the clustering variables.

dim:

The dimension of a model (K,S) defined as the number of free parmaters.

mixingProportions:

The numeric vector of the mixing proportions.

count:

A "list" of the counts of levels for each variable.

frequencies:

A "list" of the observed frequencies for each variable.

proba:

A "list" of "matrix" that contains the estimates of the levels probabilities in each clusters.

logLik:

An approximation of the maximum log-likelihood obtained by the EM algorithm.

entropy:

The entropy given by -∑_{i=1}^N∑_{k=1}^Kτ_{i,k}log≤ft(τ_{i,k}\right), where τ_{i,k} is the probability that individual i belongs to cluster k

membershipProba:

The "numeric" matrix of membership probabilities.

mapClassification:

The maximum a posteriori classification given by a vector of "integers".

Methods

==

signature(e1 = "modelKS", e2 = "ANY"): ...

[<-

signature(x = "modelKS"): ...

[

signature(x = "modelKS"): ...

is.element

signature(el = "modelKS", set = "modelKS"): ...

show

signature(object = "modelKS"): ...

simulData

signature(object = "modelKS", N = "numeric", ploidy = "numeric"): ...

read.modelKS

signature(file = "character"): ...

is.modelKS

signature(object = "modelKS"): ...

Author(s)

Wilson Toussile.

References

• Dominique Bontemps and Wilson Toussile (2013) : Clustering and variable selection for categorical multivariate data. Electronic Journal of Statistics, Volume 7, 2344-2371, ISSN.

• Wilson Toussile and Elisabeth Gassiat (2009) : Variable selection in model-based clustering using multilocus genotype data. Adv Data Anal Classif, Vol 3, number 2, 109-134.

See Also

slotNames, new, methods, showClass.

Examples

data(exModelKS)
showClass("modelKS")
slotNames("modelKS")
exModelKS
exModelKS["K"]
exModelKS["S"]
model(exModelKS)

ClustMMDD documentation built on May 2, 2019, 2:44 p.m.