SC.MEB (version 1.1)

parafun: parafun.

Description

The function parafun implements the model SC-MEB for fixed number of clusters and a sequence of beta with initial value from Gaussian mixture model

Usage

parafun(
  y,
  Adj,
  G,
  beta_grid = seq(0, 4, 0.2),
  PX = TRUE,
  maxIter_ICM = 10,
  maxIter = 50
)

Arguments

y

is n-by-d PCs.

Adj

is a sparse matrix of neighborhood.

G

is an integer specifying the numbers of clusters.

beta_grid

is a numeric vector specifying the smoothness parameter of Random Markov Field. The default is seq(0,4,0.2).

PX

is a logical value specifying the parameter expansion in EM algorithm.

maxIter_ICM

is the maximum iteration of ICM algorithm. The default is 10.

maxIter

is the maximum iteration of EM algorithm. The default is 50.

Value

a list, We briefly explain the output of the SC.MEB.

The item 'x' storing clustering results.

The item 'gam' is the posterior probability matrix.

The item 'ell' is the opposite log-likelihood.

The item 'mu' is the mean of each component.

The item 'sigma' is the variance of each component.

Details

The function parafun implements the model SC-MEB for fixed number of clusters and a sequence of beta with initial value from Gaussian mixture model

Examples

# NOT RUN {
y = matrix(rnorm(50, 0, 1), 25,2)
pos = cbind(rep(1:5, each=5), rep(1:5, 5))
Adj_sp = getneighborhood_fast(pos, 1.2)
beta_grid = c(0.5,1)
G = 2
out = parafun(y, Adj_sp, G, beta_grid)
# }