Maximum likelihood estimation in constrained parameter spaces for mixtures of factor analyzers

DOI10.1007/S11222-013-9427-ZzbMATH Open1331.62307arXiv1301.1505OpenAlexW2054668335MaRDI QIDQ5962735FDOQ5962735

Francesca Greselin, Salvatore Ingrassia

Publication date: 23 February 2016

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: Mixtures of factor analyzers are becoming more and more popular in the area of model based clustering of high-dimensional data. According to the likelihood approach in data modeling, it is well known that the unconstrained log-likelihood function may present spurious maxima and singularities and this is due to specific patterns of the estimated covariance structure, when their determinant approaches 0. To reduce such drawbacks, in this paper we introduce a procedure for the parameter estimation of mixtures of factor analyzers, which maximizes the likelihood function in a constrained parameter space. We then analyze and measure its performance, compared to the usual non-constrained approach, via some simulations and applications to real data sets.

Full work available at URL: https://arxiv.org/abs/1301.1505

Recommendations

• Maximum likelihood estimation of mixtures of factor analyzers
• Constrained monotone EM algorithms for finite mixture of multivariate Gaussians
• Eigenvalues and constraints in mixture modeling: geometric and computational issues
• Finite mixture of linear regression models: an adaptive constrained approach to maximum likelihood estimation
• The joint role of trimming and constraints in robust estimation for mixtures of Gaussian factor analyzers

zbMATH Keywords

mixture modelsmodel-based clusteringconstrained estimationfactor analyzers modeling

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cites Work

• Title not available (Why is that?)
• Finite mixture models
• Variable Selection for Model-Based Clustering
• A constrained formulation of maximum-likelihood estimation for normal mixture distributions
• The distribution of the likelihood ratio for mixtures of densities from the one-parameter exponential family
• Breakdown points for maximum likelihood estimators of location-scale mixtures
• Model-Based Gaussian and Non-Gaussian Clustering
• Estimating the components of a mixture of normal distributions
• Title not available (Why is that?)
• Extending mixtures of multivariate \(t\)-factor analyzers
• Constrained monotone EM algorithms for finite mixture of multivariate Gaussians
• A likelihood-based constrained algorithm for multivariate normal mixture models
• On the Use of Discriminant Functions in Taxonomy
• Subset Clustering of Binary Sequences, with an Application to Genomic Abnormality Data

Cited In (11)

• Robust clustering via mixtures of \(t\) factor analyzers with incomplete data
• Eigenvalues and constraints in mixture modeling: geometric and computational issues
• A finite mixture approach to joint clustering of individuals and multivariate discrete outcomes
• A data driven equivariant approach to constrained Gaussian mixture modeling
• Title not available (Why is that?)
• The joint role of trimming and constraints in robust estimation for mixtures of Gaussian factor analyzers
• Estimation of multivariate dependence structures via constrained maximum likelihood
• Maximum likelihood factor analysis with rank-deficient sample covariance matrices
• Detecting Wine Adulterations Employing Robust Mixture of Factor Analyzers
• Robust, fuzzy, and parsimonious clustering, based on mixtures of factor analyzers
• The structure of improper solutions in maximum likelihood factor analysis

Uses Software

• R
• PGMM

This page was built for publication: Maximum likelihood estimation in constrained parameter spaces for mixtures of factor analyzers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5962735)