# pgmmRJMCMC: bpgmm Model-Based Clustering Using Baysian PGMM Carries out... In bpgmm: Bayesian Model Selection Approach for Parsimonious Gaussian Mixture Models

## Description

bpgmm Model-Based Clustering Using Baysian PGMM Carries out model-based clustering using parsimonious Gaussian mixture models. MCMC are used for parameter estimation. The RJMCMC is used for model selection.

## Usage

 ```1 2 3``` ```pgmmRJMCMC(X, mInit, mVec, qnew, delta = 2, ggamma = 2, burn = 20, niter = 1000, constraint = C(0, 0, 0), dVec = c(1, 1, 1), sVec = c(1, 1, 1), Mstep = 0, Vstep = 0, SCind = 0) ```

## Arguments

 `X` the observation matrix with size p * m `mInit` the number of initial clusters `mVec` the range of the number of clusters `qnew` the number of factor for a new cluster `delta` scaler hyperparameters `ggamma` scaler hyperparameters `burn` the number of burn in iterations `niter` the number of iterations `constraint` the pgmm initial constraint, a vector of length three with binary entry. For example, c(1,1,1) means the fully constraint model `dVec` a vector of hyperparameters with length three, shape parameters for alpha1, alpha2 and bbeta respectively `sVec` sVec a vector of hyperparameters with length three, rate parameters for alpha1, alpha2 and bbeta respectively `Mstep` the indicator of whether do model selection on the number of clusters `Vstep` the indicator of whether do model selection on variance structures `SCind` the indicator of whether use split/combine step in Mstep

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30``` ```library("fabMix") library("mclust") library("pgmm") library("mvtnorm") library("mcmcse") library("MASS") library("gtools") n <- 500 p <- 10 q <- 4 K <- 10 nsim <- 10 burn <- 20 qnew <- 4 Mstep <- 1 Vstep <- 1 constraint <- c(0, 0, 0) mInit <- 20 mVec <- c(1, 20) X <- t(simData( sameLambda = TRUE, sameSigma = TRUE, K.true = K, n = n, q = q, p = p, sINV_values = 1 / ((1:p)) )\$data) pgmmRJMCMC(X, mInit, mVec, qnew, niter = nsim, burn = burn, constraint = constraint, Mstep = Mstep, Vstep = Vstep ) ```

bpgmm documentation built on July 2, 2020, 2:51 a.m.