Bounding the maximum likelihood degree

DOI10.4310/MRL.2015.V22.N6.A4MaRDI QIDQ530615zbMATH OpenOpenAlexFDO

Authors Nero Budur, Botong Wang

Publication date 10 August 2016

Published in Mathematical Research Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1411.3486

zbMATH Keywords

maximum likelihood estimationalgebraic statisticsintersection cohomologyvery affine variety

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Characterization and structure theory of statistical distributions (62E10) Applications of commutative algebra (e.g., to statistics, control theory, optimization, etc.) (13P25) Stratifications; constructible sheaves; intersection cohomology (complex-analytic aspects) (32S60) Other algebro-geometric (co)homologies (e.g., intersection, equivariant, Lawson, Deligne (co)homologies) (14F43) Topological properties in algebraic geometry (14F45) Intersection homology and cohomology in algebraic topology (55N33)

Abstract: Maximum likelihood estimation is a fundamental computational problem in statistics. In this note, we give a bound for the maximum likelihood degree of algebraic statistical models for discrete data. As usual, such models are identified with special very affine varieties. Using earlier work of Franecki and Kapranov, we prove that the maximum likelihood degree is always less or equal to the signed intersection-cohomology Euler characteristic. We construct counterexamples to a bound in terms of the usual Euler characteristic conjectured by Huh and Sturmfels.

Recommendations

• The maximum likelihood degree
• The maximum likelihood degree of mixtures of independence models
• Maximum likelihood geometry in the presence of data zeros
• Maximum likelihood for dual varieties
• The maximum likelihood degree of toric varieties

Cited in

(16)

• A note on maximized likelihood sets
• scientific article; zbMATH DE number 5494141 (Why is no real title available?)
• Maximum likelihood for dual varieties
• The maximum likelihood degree of a very affine variety
• Logarithmic cotangent bundles, Chern‐Mather classes, and the Huh‐Sturmfels involution conjecture
• Maximum likelihood degree of Fermat hypersurfaces via Euler characteristics
• The Approximate Maximum-Likelihood Certificate
• Counting and locating the solutions of polynomial systems of maximum likelihood equations. I.
• Discrete statistical models with rational maximum likelihood estimator
• Maximum likelihood degree of surjective rational maps
• The maximum likelihood data singular locus
• The maximum likelihood degree
• Data-discriminants of likelihood equations
• Maximum-Likelihood Estimation, the CramÉr–Rao Bound, and the Method of Scoring With Parameter Constraints
• The maximum likelihood degree of mixtures of independence models
• Likelihood equations and scattering amplitudes

This page was built for publication: Bounding the maximum likelihood degree

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q530615)