Learn R Programming

bayesMCClust (version 1.0)

Mixtures-of-Experts Markov Chain Clustering and Dirichlet Multinomial Clustering

Description

This package provides various Markov Chain Monte Carlo (MCMC) sampler for model-based clustering of discrete-valued time series obtained by observing a categorical variable with several states (in a Bayesian approach). In order to analyze group membership, we provide also an extension to the approaches by formulating a probabilistic model for the latent group indicators within the Bayesian classification rule using a multinomial logit model.

Copy Link

Version

Install

install.packages('bayesMCClust')

Monthly Downloads

3

Version

1.0

License

GPL-2

Maintainer

Christoph Pamminger

Last Published

January 31st, 2012

Functions in bayesMCClust (1.0)

Calculates the Posterior Expectation of the Cluster-Specific Parameter Matrices (only for DMC[Ext])

Calculates Inefficiency Factors of the MCMC Draws Obtained for the Cluster-Specific Parameters

Calculates Posterior Expectations, Standard Deviations and (Optionally) HPD Intervals for the MNL Regression Coefficients

A Small MCC/DMC Example Data Set

plotTypicalMembers

Plots Time Series of 'Typical' Group Members

bayesMCClust-package

Mixtures-of-Experts Markov Chain Clustering and Dirichlet Multinomial Clustering

Produces Balloon Plots and LaTeX-Style Tables of the Transition Matrices

plotLikeliPaths

Plots Paths of Likelihoods And (Prior) Densities

calcSegmentationPower

Calculates the 'Segmentation Power' of the Specified Classification

Calculates Model Selection Criteria For Several (Independent) MCMC Runs And Various Numbers $H$ of Clusters

calcVariationDMC

Analyses How Much Unobserved Heterogeneity Is Present in the Various Clusters by Computing the Within-Group Variability of the Cluster-Specific Transition Parameters of DMC

transformDataToNjki

Transform Markov Chain (Time Series) Data Into Transition Frequency Structure

Bayesian Multinomial Logit Regression Using Auxiliary Mixture Sampling

Calculates the Posterior Expectation and Standard Deviations of the Average Cluster-Specific Transition Matrices

calcLongRunDist

Calculates And Plots the Long-Run Distribution Over the Categories of the Outcome Variable After Certain Periods.

Markov Chain Clustering With And Without Mixtures-of-Experts Extension

Dirichlet Multinomial Clustering With And Without Mixtures-of-Experts Extension

Produces Scatter Plots of MCMC Draws

calcAllocations

Computes Group Sizes, Group Membership and Individual Posterior Classification Probabilities

Calculates (And Plots) the Stationary Distribution (Steady State)

Calculates the Entropy of a Given Classification

MCCExtExampleData

An Extended MCC/DMC Example Data Set Including Covariates

LMEntryPaperData

Data From Fruehwirth-Schnatter et al. (2011): "Labor market entry and earnings dynamics: Bayesian inference using mixtures-of-experts Markov chain clustering"