cluster_mod: Estimate the optimal posterior distribution of the data...
In xiangli2pro/hbcm: HBCM model for cluster analysis

heterogbcm

R Documentation

Estimate the optimal posterior distribution of the data column labels.

Description

'heterogbcm( )' gives the optimal posterior distribution of the labels, which can be used to derive the optimal label assignment of the data columns.

'get_random_qc( )' assign initial prob between 0 and 1 to cluster matrix

'get_prob( )' get num_prob with sum equal to total_prob

'heterogbcm_prob_qc( )' set initial distribution of label to be probability rather than discrete value.

'heterogbcm_converge_qc( )' set initial distribution of label to be probability rather than discrete value.

'obj_logL' calculates the negative log-likelihood function of data fitted into the HBCM model

Usage

heterogbcm(x, centers, tol, iter, iter_init = 3, labels, verbose = FALSE)

get_random_qc(grp, centers)

get_prob(num_prob, total_prob)

heterogbcm_prob_qc(
  x,
  centers,
  tol,
  iter,
  iter_init = 3,
  labels,
  verbose = FALSE
)

heterogbcm_converge_qc(
  x,
  centers,
  tol,
  iter,
  iter_init = 3,
  labels,
  verbose = FALSE
)

obj_qc(x, centers, ppi, omega, qalpha, hlambda, hsigma)

obj_qalpha(x, centers, omega, qc, hlambda, hsigma)

obj_ppi(centers, qc)

obj_omega(centers, qalpha)

obj_hlambda(x, centers, qc, qalpha)

obj_hsigma(x, centers, qc, qalpha, hlambda)

obj_logL(x, centers, ppi, omega, qc, qalpha, hlambda, hsigma)

Arguments

`x`	a numeric matrix data.
`centers`	an integer specifying the number of clusters.
`tol`	numerical tolerance of the iteration updates.
`iter`	number of iterations.
`iter_init`	number of iterations of parameters initial estimation, default is 3.
`labels`	a vector specifying the cluster labels of the columns of x.
`verbose`	if TRUE, print iteration information.
`ppi`	probability of multi-nulli distribution.
`omega`	group-correlation matrix.
`qalpha`	posterior distribution of parameter vector alpha.
`hlambda`	heterogeneous parameter vector Lambda.
`hsigma`	heterogeneous parameter vector Sigma.
`qc`	posterior distribution of labels.

Value

A list of values.

`omega`	estimated optimal group-correlation matrix.
`hlambda`	estimated optimal heterogeneous parameter Lambda.
`hsigma`	estimated optimal heterogeneous parameter Sigma.
`obj_logL_val`	vector of -logL from each iteration.
`qalpha`	estimated optimal posterior distribution of the alpha.
`qc`	estimated optimal posterior distribution of the column labels.
`cluster`	a vector of integers (from 1:k) indicating the cluster to which each column is allocated.