MclustSBICupdate: SBIC Update Function for the BIC Based Mclust

In radeksalac/mclustSBIC: Bayesian Information Criterion for Singular Gaussian Mixture Models

Description

Updates the 'Mclust' output such that the optimal model is selected in accordance to the Singular BIC (sBIC) criterion. The class remains unchanged so that other functions and methods from package mclust can be applied. Note, that some details such as plot titles or axis labels might not be adjusted.

Usage

MclustSBICupdate(object,...)

Arguments

object

An object of class 'Mclust' resulting from a call to Mclust.

Value

Return an object of class 'Mclust' providing the optimal (according to sBIC) mixture model estimation.

References

Scrucca L., Fop M., Murphy T. B. and Raftery A. E. (2016) mclust 5: clustering, classification and density estimation using Gaussian finite mixture models, The R Journal, 8/1, pp. 289-317.

Drton M. and Plummer M. (2017) A Bayesian information criterion for singular models, Journal of the Royal Statistical Society, Series B, 79, Part 2, pp. 323-380.

Fraley C. and Raftery A. E. (2002) Model-based clustering, discriminant analysis and density estimation, Journal of the American Statistical Association, 97/458, pp. 611-631.

Fraley C., Raftery A. E., Murphy T. B. and Scrucca L. (2012) mclust Version 4 for R: Normal Mixture Modeling for Model-Based Clustering, Classification, and Density Estimation. Technical Report No. 597, Department of Statistics, University of Washington.

See Also

mclustSBIC, mclustMaxLik, mclustSBIClearnCoeff, nMclustParamsComp, nMclustParamsShared, priorControl, emControl, mclustModel, summary.mclustSBIC, hc, me, mclustModelNames, mclust.options

Examples

X <- iris[,-length(iris)]
b <- mclustBIC(X)
modelBIC <- Mclust(X, x = b)
plot(modelBIC, what="BIC")

modelSBIC <- MclustSBICupdate(modelBIC)
# now plot the sBIC values
# note, using the class 'Mclust' implies the y-axis label 'BIC'
# however, since MclustSBICupdate was applied, the plotted values truly are the sBIC
plot(modelSBIC, what="BIC")

modDR = MclustDR(modelSBIC)
plot(modDR,what = "classification") 

radeksalac/mclustSBIC documentation built on Jan. 15, 2022, 8:12 a.m.