Probability matrices, non-negative rank, and parameterization of mixture models

DOI10.1016/J.LAA.2010.03.010MaRDI QIDQ975609zbMATH OpenOpenAlexFDO

Authors Enrico Carlini, Fabio Rapallo

Publication date 10 June 2010

Published in Linear Algebra and its Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0911.0412

zbMATH Keywords

mixture modelscontingency tablesparameter redundancyoptimization on algebraic varieties

Mathematics Subject Classification ID

Contingency tables (62H17) Stochastic matrices (15B51)

Abstract: In this paper we parameterize non-negative matrices of sum one and rank at most two. More precisely, we give a family of parameterizations using the least possible number of parameters. We also show how these parameterizations relate to a class of statistical models, known in Probability and Statistics as mixture models for contingency tables.

Recommendations

• Perturbation of matrices and nonnegative rank with a view toward statistical models
• The nonnegative rank factorizations of nonnegative matrices
• Fixed points of the EM algorithm and nonnegative rank boundaries
• Maximum likelihood for matrices with rank constraints
• Parameterizing mixture models with generalized moments

Cites work

• scientific article; zbMATH DE number 1817585 (Why is no real title available?)
• scientific article; zbMATH DE number 3940389 (Why is no real title available?)
• scientific article; zbMATH DE number 52497 (Why is no real title available?)
• Algebraic Statistics for Computational Biology
• Detecting parameter redundancy
• Miscellanea. Estimation in parameter-redundant models
• Non-negative matrix factorization with fixed row and column sums
• Nonnegative matrix factorization and I-divergence alternating minimization
• Nonnegative ranks, decompositions, and factorizations of nonnegative matrices
• On reduced rank nonnegative matrix factorization for symmetric nonnegative matrices
• Open Problems in Algebraic Statistics

Cited in

(8)

• Rank-to-rank embeddings, Laver tables, and Galton-Watson processes
• Matrix inference and estimation in multi-layer models*
• Moment matrices: Applications in mixtures
• Heuristics for exact nonnegative matrix factorization
• On the geometric interpretation of the nonnegative rank
• Hierarchical low-rank structure of parameterized distributions
• Perturbation of matrices and nonnegative rank with a view toward statistical models
• Rank-one corrections of nonnegative matrices, with an application to matrix population models

This page was built for publication: Probability matrices, non-negative rank, and parameterization of mixture models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q975609)