Estimating the number of components in finite mixture models via the group-sort-fuse procedure

DOI10.1214/21-AOS2072zbMath1486.62062arXiv2005.11641OpenAlexW3027727313MaRDI QIDQ2073694

Abbas Khalili, Tudor Manole

Publication date: 7 February 2022

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2005.11641

zbMATH Keywords

Wasserstein distancefinite mixture modelsstrong identifiabilitymaximum penalized likelihood estimation

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Estimation in multivariate analysis (62H12) Point estimation (62F10)

Related Items

GroupSortFuse ⋮ Determine the number of clusters by data augmentation ⋮ Prices, profits, proxies, and production ⋮ Order selection for regression-based hidden Markov model ⋮ Statistical inference for normal mixtures with unknown number of components

Uses Software

• mclust
• R
• mixtools
• ElemStatLearn

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Asymptotic Behaviour of the Posterior Distribution in Overfitted Mixture Models
• Nearly unbiased variable selection under minimax concave penalty
• The Adaptive Lasso and Its Oracle Properties
• Convergence rates of parameter estimation for some weakly identifiable finite mixtures
• Hypothesis test for normal mixture models: the EM approach
• One-step sparse estimates in nonconcave penalized likelihood models
• Consistent estimation of a mixing distribution
• Estimating the dimension of a model
• Asymptotics for likelihood ratio tests under loss of identifiability
• Introductory lectures on convex optimization. A basic course.
• Testing the order of a model using locally conic parametrization: Population mixtures and stationary ARMA processes
• Bayesian analysis of mixture models with an unknown number of components\,--\,an alternative to reversible jump methods.
• Rates of convergence for the Gaussian mixture sieve.
• Convergence rates for density estimation with Bernstein polynomials.
• Consistent estimation of mixture complexity.
• Optimal rate of convergence for finite mixture models
• Probability inequalities for likelihood ratios and convergence rates of sieve MLEs
• Convergence of latent mixing measures in finite and infinite mixture models
• MCLUST: Software for model-based cluster analysis
• On posterior contraction of parameters and interpretability in Bayesian mixture modeling
• Strong identifiability and optimal minimax rates for finite mixture estimation
• Optimal estimation of Gaussian mixtures via denoised method of moments
• Robust estimation of mixing measures in finite mixture models
• Finite mixture and Markov switching models.
• A finite mixture distribution for modelling multinomial extra variation
• Bayesian Repulsive Gaussian Mixture Model
• Robust Estimation of Mixture Complexity
• Non-finite Fisher information and homogeneity: an EM approach
• Finite Mixture Models with Concomitant Information: Assessing Diagnostic Criteria for Diabetes
• Bayesian Model Selection in Finite Mixtures by Marginal Density Decompositions
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Testing for a Finite Mixture Model with Two Components
• Mixture Models With a Prior on the Number of Components
• Penalized minimum‐distance estimates in finite mixture models
• Regularization Parameter Selections via Generalized Information Criterion
• Testing the Order of a Finite Mixture
• A Bayesian Information Criterion for Singular Models
• Hidden Markov Models With Applications in Cell Adhesion Experiments
• Order Selection in Finite Mixture Models With a Nonsmooth Penalty
• Identifiability of Finite Mixtures
• On strong identifiability and convergence rates of parameter estimation in finite mixtures
• A new look at the statistical model identification