RmixmodCombi (version 1.0)

MixmodCombi-class: [MixmodCombi] class

Description

Class of an output from the mixmodCombi function

Details

mixmodOutput

[MixmodCluster] object. EM/BIC solution from which the hierarchy is computed. Either provided by the user or computed by a call to the mixmodCluster function

hierarchy

list of MixmodCombiSol objects, each of which is the solution for the corresponding number of clusters obtained by hierarchically combining the EM/BIC solution according to the method proposed in the article in the references. Each one contains: the number of clusters, the partition of the data, the posterior probabilities of each class for each observation, the entropy value for the solution and a "combining matrix" combiM which enables to get the K-cluster solution from the (K+1)-cluster solution (please see the combMat function documentation about the combining matrices and how to use them).

ICLNbCluster

number of clusters selected by ICL, according to the mixmodOutput solution (if the criterion option has not been changed when calling the mixmodCombi function).

References

J.-P. Baudry, A. E. Raftery, G. Celeux, K. Lo and R. Gottardo (2010). Combining mixture components for clustering. Journal of Computational and Graphical Statistics, 19(2):332-353.

Examples

set.seed(1)

data(Baudry_etal_2010_JCGS_examples)

res <- mixmodCombi(ex4.1, nbCluster = 1:8)

res # is of class MixmodCombi

getSlots("MixmodCombi")