gibbSampling: Gibbs sampling algorithm

Description Usage Arguments Value

Description

This computes the Gibbs sampling procedure

Usage

1
gibbSampling(K, ipdata, thr = Inf, burnIn, numIter, constr, mix, niwpr)

Arguments

K

maximum number of clusters

ipdata

data matrix where columns are individual observations and rows are dimensions

thr

maximum number of rejections per observations

burnIn

number of burn-in for the chain. Must be less than number of iterations

numIter

total number of iterations

constr

function for the constraints

mix

one of "tmog" and "motg"

niwpr

list of NIW prior parameters

Value

a list of results consisting of weights, mean, and covariance matrix for every iterations, along with the rejected proposals


pasudyan/ConstrMixMod documentation built on May 10, 2019, 8:26 a.m.