gmmsslm {gmmsslm}

Fitting Gaussian mixture model to a complete classified dataset or an incomplete classified dataset with/without the missing-data mechanism.

Description

Fitting Gaussian mixture model to a complete classified dataset or an incomplete classified dataset with/without the missing-data mechanism.

Usage

gmmsslm(
  dat,
  zm,
  pi = NULL,
  mu = NULL,
  sigma = NULL,
  paralist = NULL,
  xi = NULL,
  type,
  iter.max = 500,
  eval.max = 500,
  rel.tol = 1e-15,
  sing.tol = 1e-15
)

Arguments

dat

An n\times p matrix where each row represents an individual observation

zm

An n-dimensional vector containing the class labels including the missing-label denoted as NA.

pi

A g-dimensional vector for the initial values of the mixing proportions.

mu

A p \times g matrix for the initial values of the location parameters.

sigma

A p\times p covariance matrix,or a list of g covariance matrices with dimension p\times p \times g. It is assumed to fit the model with a common covariance matrix if sigma is a p\times p covariance matrix; otherwise it is assumed to fit the model with unequal covariance matrices.

paralist

A list containing the required parameters (\pi, \mu, \Sigma).

xi

A 2-dimensional vector containing the initial values of the coefficients in the logistic function of the Shannon entropy.

type

Three types of Gaussian mixture models, 'ign' indicates fitting the model to a partially classified sample on the basis of the likelihood that ignores the missing label mechanism, 'full' indicates fitting the model to a partially classified sample on the basis of the full likelihood, taking into account the missing-label mechanism, and 'com' indicate fitting the model to a completed classified sample.

iter.max

Maximum number of iterations allowed. Defaults to 500

eval.max

Maximum number of evaluations of the objective function allowed. Defaults to 500

rel.tol

Relative tolerance. Defaults to 1e-15

sing.tol

Singular convergence tolerance; defaults to 1e-20.

Value

A gmmsslmFit object containing the following slots:

objective

Value of objective likelihood

convergence

Value of convergence

iteration

Number of iterations

obs

Input data matrix

n

Number of observations

p

Number of variables

g

Number of Gaussian components

type

Type of Gaussian mixture model

pi

Estimated vector of the mixing proportions

mu

Estimated matrix of the location parameters

sigma

Estimated covariance matrix or list of covariance matrices

xi

Estimated coefficient vector for a logistic function of the Shannon entropy

Examples

n<-150
pi<-c(0.25,0.25,0.25,0.25)
sigma<-array(0,dim=c(3,3,4))
sigma[,,1]<-diag(1,3)
sigma[,,2]<-diag(2,3)
sigma[,,3]<-diag(3,3)
sigma[,,4]<-diag(4,3)
mu<-matrix(c(0.2,0.3,0.4,0.2,0.7,0.6,0.1,0.7,1.6,0.2,1.7,0.6),3,4)
dat<-rmix(n=n,pi=pi,mu=mu,sigma=sigma)
xi<-c(-0.5,1)
m<-rlabel(dat=dat$Y,pi=pi,mu=mu,sigma=sigma,xi=xi)
zm<-dat$clust
zm[m==1]<-NA
inits<-initialvalue(g=4,zm=zm,dat=dat$Y)

fit_pc<-gmmsslm(dat=dat$Y,zm=zm,paralist=inits,xi=xi,type='full')