Uniqueness of estimation and identifiability in mixture models

DOI10.2307/3315807zbMath0779.62032OpenAlexW2084676098WikidataQ58047137 ScholiaQ58047137MaRDI QIDQ4272579

Bruce G. Lindsay, Kathryn Roeder

Publication date: 19 January 1994

Published in: Canadian Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.2307/3315807

zbMATH Keywords

mixing distributionidentifiabilitysufficient conditions for uniquenessnonparametric maximum-likelihood estimatormixture model spacetotally positive kernels

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Existence theories in calculus of variations and optimal control (49J99)

Related Items (19)

Mixture models: building a parameter space ⋮ Equivalence of Truncated Count Mixture Distributions and Mixtures of Truncated Count Distributions ⋮ Moment-based oscillation properties of mixture models ⋮ A review of reliable maximum likelihood algorithms for semiparametric mixture models ⋮ A review of semiparametric mixture models ⋮ Almost sure uniqueness of a global minimum without convexity ⋮ A general approach to full-range tail dependence copulas ⋮ Nonparametric estimation of convex models via mixtures ⋮ A nonparametric mixture approach to density and null proportion estimation in large‐scale multiple comparison problems ⋮ Fisher-Pitman Permutation Tests Based on Nonparametric Poisson Mixtures with Application to Single Cell Genomics ⋮ Partially projected gradient algorithms for computing nonparametric maximum likelihood estimates of mixing distributions ⋮ Ordinal aggregation results via Karlin's variation diminishing property ⋮ Uniqueness of the maximum likelihood estimator for 𝑘-monotone densities ⋮ Exploring Extra-Binomial Variation in Teratology Data Using Continuous Mixtures ⋮ A linearization procedure and a VDM/ECM algorithm for penalized and constrained nonparametric maximum likelihood estimation for mixture models ⋮ Mixed Poisson Distributions ⋮ Mixture Models and the Krätzel Integral Transform ⋮ Convergence of the EM algorithm for continuous mixing distributions ⋮ Parameterizing mixture models with generalized moments

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• The geometry of mixture likelihoods, part II: The exponential family
• The geometry of mixture likelihoods: A general theory
• Nonparametric maximum likelihood estimation by the method of sieves
• Mixtures of exponential distributions
• Maximum likelihood estimation of a compound Poisson process
• Identifiability of mixtures of exponential families
• On the Mixture of Distributions
• Semiparametric Estimation in the Rasch Model and Related Exponential Response Models, Including a Simple Latent Class Model for Item Analysis
• Residual Diagnostics for Mixture Models
• Nonparametric Maximum Likelihood Estimation of a Mixing Distribution
• Identifiability of Finite Mixtures
• Identifiability of Mixtures

This page was built for publication: Uniqueness of estimation and identifiability in mixture models