Mixtures of generalized hyperbolic distributions and mixtures of skew-t distributions for model-based clustering with incomplete data

DOI10.1016/J.CSDA.2018.08.016MaRDI QIDQ1799870zbMATH OpenOpenAlexWikidataFDO

Authors Yuhong Wei, Yang Tang, Paul D. McNicholas

Publication date 19 October 2018

Published in Computational Statistics and Data Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1703.02177

zbMATH Keywords

clusteringmissing datamixture modelsskew-\(t\)generalized hyperbolic

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Abstract: Robust clustering from incomplete data is an important topic because, in many practical situations, real data sets are heavy-tailed, asymmetric, and/or have arbitrary patterns of missing observations. Flexible methods and algorithms for model-based clustering are presented via mixture of the generalized hyperbolic distributions and its limiting case, the mixture of multivariate skew-t distributions. An analytically feasible EM algorithm is formulated for parameter estimation and imputation of missing values for mixture models employing missing at random mechanisms. The proposed methodologies are investigated through a simulation study with varying proportions of synthetic missing values and illustrated using a real dataset. Comparisons are made with those obtained from the traditional mixture of generalized hyperbolic distribution counterparts by filling in the missing data using the mean imputation method.

Recommendations

• Robust model-based clustering via mixtures of skew-t distributions with missing information
• Supervised learning of multivariate skew normal mixture models with missing information
• A mixture of generalized hyperbolic distributions
• Robust statistical modelling using the multivariate skew \(t\) distribution with complete and incomplete data
• Robust mixture modeling based on scale mixtures of skew-normal distributions

Cites work

• scientific article; zbMATH DE number 4088699 (Why is no real title available?)
• scientific article; zbMATH DE number 3770749 (Why is no real title available?)
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 3599301 (Why is no real title available?)
• scientific article; zbMATH DE number 3600847 (Why is no real title available?)
• scientific article; zbMATH DE number 2231189 (Why is no real title available?)
• A general class of multivariate skew-elliptical distributions
• A mixture of generalized hyperbolic distributions
• A mixture of generalized hyperbolic factor analyzers
• A new class of multivariate skew distributions with applications to bayesian regression models
• Analysis of multivariate skew normal models with incomplete data
• Analytic calculations for the EM algorithm for multivariate skew-\(t\) mixture models
• Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy
• Capturing patterns via parsimonious \(t\) mixture models
• Clustering with the multivariate normal inverse Gaussian distribution
• Clustering, classification, discriminant analysis, and dimension reduction via generalized hyperbolic mixtures
• Computationally efficient learning of multivariate \(t\) mixture models with missing information
• Estimating the dimension of a model
• Extending mixtures of multivariate \(t\)-factor analyzers
• Finite mixtures of multivariate skew \(t\)-distributions: some recent and new results
• Fractionally-supervised classification
• Hidden truncation hyperbolic distributions, finite mixtures thereof, and their application for clustering
• High-dimensional data clustering
• How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis
• Inference and missing data
• Infinite divisibility of the hyperbolic and generalized inverse Gaussian distributions
• Learning from incomplete data via parameterized \(t\) mixture models through eigenvalue decomposition
• Mixture models: theory, geometry and applications
• Mixtures of multivariate power exponential distributions
• Mixtures of skew-\(t\) factor analyzers
• Model-Based Gaussian and Non-Gaussian Clustering
• Model-based classification via mixtures of multivariate \(t\)-factor analyzers
• Model-based clustering
• Model-based clustering, classification, and discriminant analysis via mixtures of multivariate \(t\)-distributions
• Modelling high-dimensional data by mixtures of factor analyzers
• Multivariate extended skew-\(t\) distributions and related families
• On fast supervised learning for normal mixture models with missing information
• Parameter expansion to accelerate EM: the PX-EM algorithm
• Robust model-based clustering via mixtures of skew-\(t\) distributions with missing information
• Robust statistical modelling using the multivariate skew \(t\) distribution with complete and incomplete data
• Self-decomposability of the generalized inverse Gaussian and hyperbolic distributions
• Serial and parallel implementations of model-based clustering via parsimonious Gaussian mixture models
• Statistical properties of the generalized inverse Gaussian distribution
• THE POPULATION FREQUENCIES OF SPECIES AND THE ESTIMATION OF POPULATION PARAMETERS
• The distribution of the likelihood ratio for mixtures of densities from the one-parameter exponential family

Cited in

(18)

• Learning from incomplete data via parameterized \(t\) mixture models through eigenvalue decomposition
• Infinite mixtures of multivariate normal-inverse Gaussian distributions for clustering of skewed data
• Robust statistical modelling using the multivariate skew \(t\) distribution with complete and incomplete data
• Model-based clustering using a new multivariate skew distribution
• A comparative study of methods for estimating model-agnostic Shapley value explanations
• Missing values and directional outlier detection in model-based clustering
• Fast model-based clustering of partial records
• Likelihood-based inference for linear mixed-effects models using the generalized hyperbolic distribution
• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Soft clustering
• Model-based clustering with missing not at random data
• Asymmetric clusters and outliers: mixtures of multivariate contaminated shifted asymmetric Laplace distributions
• Parsimonious mixture of mean-mixture of normal distributions with missing data
• Some properties of the multivariate generalized hyperbolic laws
• Robust model-based clustering via mixtures of skew-\(t\) distributions with missing information
• Supervised learning of multivariate skew normal mixture models with missing information
• A mixture of coalesced generalized hyperbolic distributions
• Model-based clustering and outlier detection with missing data

Describes a project that uses

Uses Software

• QRM

This page was built for publication: Mixtures of generalized hyperbolic distributions and mixtures of skew-\(t\) distributions for model-based clustering with incomplete data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1799870)