Parsimonious mixtures of multivariate contaminated normal distributions

DOI10.1002/BIMJ.201500144zbMATH Open1353.62124arXiv1305.4669OpenAlexW2963702653WikidataQ39503075 ScholiaQ39503075MaRDI QIDQ2833487FDOQ2833487

Authors: Antonio Punzo, Paul D. McNicholas 

Publication date: 18 November 2016

Published in: Biometrical Journal (Search for Journal in Brave)

Abstract: A mixture of multivariate contaminated normal distributions is developed for model-based clustering. In addition to the parameters of the classical normal mixture, our contaminated mixture has, for each cluster, a parameter controlling the proportion of mild outliers and one specifying the degree of contamination. Crucially, these parameters do not have to be specified a priori, adding a flexibility to our approach. Parsimony is introduced via eigen-decomposition of the component covariance matrices, and sufficient conditions for the identifiability of all the members of the resulting family are provided. An expectation-conditional maximization algorithm is outlined for parameter estimation and various implementation issues are discussed. Using a large scale simulation study, the behaviour of the proposed approach is investigated and comparison with well-established finite mixtures is provided. The performance of this novel family of models is also illustrated on artificial and real data.

Full work available at URL: https://arxiv.org/abs/1305.4669

Recommendations

• Asymmetric clusters and outliers: mixtures of multivariate contaminated shifted asymmetric Laplace distributions
• Mixtures of multivariate contaminated normal regression models
• Model-based clustering via new parsimonious mixtures of heavy-tailed distributions
• Cluster analysis of a mixture of multivariate probability distributions under uniform ``contamination
• Clustering asymmetrical data with outliers: parsimonious mixtures of contaminated mean-mixture of normal distributions

zbMATH Keywords

contaminated normal distributionEM algorithmmixture modelsmodel-based clusteringmultivariateparsimonycovariance matricescontamination

Mathematics Subject Classification ID

Point estimation (62F10) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Cites Work

• Robust fitting of mixture regression models
• Estimating the dimension of a model
• Title not available (Why is that?)
• Finite mixture models
• Serial and parallel implementations of model-based clustering via parsimonious Gaussian mixture models
• The distribution of the likelihood ratio for mixtures of densities from the one-parameter exponential family
• A robust method for cluster analysis
• Breakdown points for maximum likelihood estimators of location-scale mixtures
• Maximum likelihood estimation via the ECM algorithm: A general framework
• How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis
• Model-Based Gaussian and Non-Gaussian Clustering
• Dealing With Label Switching in Mixture Models
• The Masking Breakdown Point of Multivariate Outlier Identification Rules
• Computational and Inferential Difficulties with Mixture Posterior Distributions
• Title not available (Why is that?)
• Mixture Model-Based Classification
• On the Identifiability of Finite Mixtures
• A general trimming approach to robust cluster analysis
• Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models
• Choosing initial values for the EM algorithm for finite mixtures
• Trimmed \(k\)-means: An attempt to robustify quantizers
• Title not available (Why is that?)
• An Algorithm for Simultaneous Orthogonal Transformation of Several Positive Definite Symmetric Matrices to Nearly Diagonal Form
• Outlier analysis.
• Model-based clustering, classification, and discriminant analysis via mixtures of multivariate \(t\)-distributions
• Model-based classification using latent Gaussian mixture models
• Title not available (Why is that?)
• Model based labeling for mixture models
• Constrained monotone EM algorithms for finite mixture of multivariate Gaussians
• A likelihood-based constrained algorithm for multivariate normal mixture models
• Title not available (Why is that?)
• Identifiability of Finite Mixtures of Elliptical Distributions
• Fixed point clusters for linear regression: Computation and comparison
• Robust Estimation of the Mean and Covariance Matrix from Data with Missing Values
• Mixture Models, Robustness, and the Weighted Likelihood Methodology
• Maximum likelihood estimation of heterogeneous mixtures of Gaussian and uniform distributions
• Nearest-Neighbor Clutter Removal for Estimating Features in Spatial Point Processes
• A mixture of mixture models for a classification problem: the unity measure error
• A review of robust clustering methods
• Robustness Properties of k Means and Trimmed k Means
• The influence function of the TCLUST robust clustering procedure
• Finite mixtures of unimodal beta and gamma densities and the \(k\)-bumps algorithm
• Identifiability of Finite Mixtures
• A note on the maximum deviation of the scale-contaminated normal to the best normal distribution
• Statistical theory in clustering
• Testing the number of components in a normal mixture
• Robust mixture regression using the \(t\)-distribution
• Title not available (Why is that?)
• On the breakdown behavior of the TCLUST clustering procedure
• The Identification of Multiple Outliers
• Mixture Models, Outliers, and the EM Algorithm
• Degeneracy of the EM algorithm for the MLE of multivariate Gaussian mixtures and dynamic constraints
• An Application of the Laplace Method to Finite Mixture Distributions
• Degeneracy in the maximum likelihood estimation of univariate Gaussian mixtures with EM.
• Likelihood ratio tests of the number of components in a normal mixture with unequal variances

Cited In (56)

• Contamination transformation matrix mixture modeling for skewed data groups with heavy tails and scatter
• Flexible mixture regression with the generalized hyperbolic distribution
• A Laplace-based model with flexible tail behavior
• Modeling frequency and severity of claims with the zero-inflated generalized cluster-weighted models
• Multiple scaled symmetric distributions in allometric studies
• A new look at the inverse Gaussian distribution with applications to insurance and economic data
• Asymmetric clusters and outliers: mixtures of multivariate contaminated shifted asymmetric Laplace distributions
• Mixtures of multivariate contaminated normal regression models
• Transformation mixture modeling for skewed data groups with heavy tails and scatter
• Model-based clustering
• Multivariate Contaminated Normal Censored Regression Model: Properties and Maximum Likelihood Inference
• On model-based clustering of directional data with heavy tails
• Dichotomous unimodal compound models: application to the distribution of insurance losses
• Maximum a-posteriori estimation of autoregressive processes based on finite mixtures of scale-mixtures of skew-normal distributions
• Unconstrained representation of orthogonal matrices with application to common principal components
• Robust model-based clustering with mild and gross outliers
• Multivariate response and parsimony for Gaussian cluster-weighted models
• Minimum profile Hellinger distance estimation for a semiparametric mixture model
• On the inevitability of a paradox in shrinkage estimation for scale mixtures of normals.
• Outlier detection in multivariate functional data through a contaminated mixture model
• Model-based clustering with determinant-and-shape constraint
• Finding outliers in Gaussian model-based clustering
• Extending finite mixtures of \(t\) linear mixed-effects models with concomitant covariates
• Dimension-wise scaled normal mixtures with application to finance and biometry
• Modeling the cryptocurrency return distribution via Laplace scale mixtures
• Model-based clustering via new parsimonious mixtures of heavy-tailed distributions
• Estimation and computations for Gaussian mixtures with uniform noise under separation constraints
• Clustering asymmetrical data with outliers: parsimonious mixtures of contaminated mean-mixture of normal distributions
• CO\(_2\) emissions and growth: a bivariate bidimensional mean-variance random effects model
• Model-based time-varying clustering of multivariate longitudinal data with covariates and outliers
• Hypothesis Testing for Mixture Model Selection
• Bayesian analysis for mixture of latent variable hidden Markov models with multivariate longitudinal data
• Robust fitting of mixture models using weighted complete estimating equations
• Robust inference for parsimonious model-based clustering
• Gaussian parsimonious clustering models with covariates and a noise component
• Multiple scaled contaminated normal distribution and its application in clustering
• Model-based clustering and outlier detection with missing data
• Fitting insurance and economic data with outliers: a flexible approach based on finite mixtures of contaminated gamma distributions
• Multivariate hidden Markov regression models: random covariates and heavy-tailed distributions
• The multivariate tail-inflated normal distribution and its application in finance
• Mixtures of Matrix-Variate Contaminated Normal Distributions
• On latent idealized models in symbolic datasets: unveiling signals in noisy sequencing data
• Contaminated Kent mixture model for clustering non-spherical directional data with heavy tails or scatter
• Robust clustering in regression analysis via the contaminated Gaussian cluster-weighted model
• Unsupervised classification with a family of parsimonious contaminated shifted asymmetric Laplace mixtures
• Cluster analysis of a mixture of multivariate probability distributions under uniform ``contamination
• Conditional mixture modelling for heavy-tailed and skewed data
• A time-varying multivariate noncentral contaminated normal copula model and its application to the visualized dependence analysis of Hong Kong stock markets
• A simulation study comparing model fit measures of structural equation modeling with multivariate contaminated normal distribution
• Robust clustering based on trimming
• Parsimonious mixture of mean-mixture of normal distributions with missing data
• Model-based clustering of functional data via mixtures of \(t\) distributions
• Parsimony and parameter estimation for mixtures of multivariate leptokurtic-normal distributions
• Matrix-variate hidden Markov regression models: fixed and random covariates
• Missing values and directional outlier detection in model-based clustering
• Skew multiple scaled mixtures of normal distributions with flexible tail behavior and their application to clustering

Uses Software

• mclust
• R
• gclus
• PARVUS
• Rmixmod
• mixture
• otrimle

This page was built for publication: Parsimonious mixtures of multivariate contaminated normal distributions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2833487)