Toric invariant theory for maximum likelihood estimation in log-linear models

DOI10.2140/ASTAT.2021.12.187zbMATH Open1487.14100arXiv2012.07793OpenAlexW3111638709MaRDI QIDQ2076293FDOQ2076293

Authors: Kathlén Kohn, Philipp Reichenbach, Anna Leah Seigal, Carlos Enrique Améndola Cerón 

Publication date: 16 February 2022

Published in: Algebraic Statistics (Search for Journal in Brave)

Abstract: We establish connections between invariant theory and maximum likelihood estimation for discrete statistical models. We show that norm minimization over a torus orbit is equivalent to maximum likelihood estimation in log-linear models. We use notions of stability under a torus action to characterize the existence of the maximum likelihood estimate, and discuss connections to scaling algorithms.

Full work available at URL: https://arxiv.org/abs/2012.07793

Recommendations

• Invariant Theory and Scaling Algorithms for Maximum Likelihood Estimation
• The maximum likelihood degree of toric varieties
• The maximum likelihood degree
• Maximum Likelihood Estimation from a Tropical and a Bernstein–Sato Perspective
• Maximum likelihood estimation of toric Fano varieties

zbMATH Keywords

log-linear modelsmaximum likelihood estimationgraphical modelstorus actionsnull conescaling algorithms

Mathematics Subject Classification ID

Point estimation (62F10) Probabilistic graphical models (62H22) Algebraic statistics (62R01) Geometric invariant theory (14L24) Applications of linear algebraic groups to the sciences (20G45)

Cites Work

• Concerning nonnegative matrices and doubly stochastic matrices
• A Relationship Between Arbitrary Positive Matrices and Doubly Stochastic Matrices
• On a Least Squares Adjustment of a Sampled Frequency Table When the Expected Marginal Totals are Known
• Title not available (Why is that?)
• Maximum likelihood estimation in log-linear models
• On the toric algebra of graphical models
• Title not available (Why is that?)
• Generalized Iterative Scaling for Log-Linear Models
• Convexity and Commuting Hamiltonians
• Geometric Invariant Theory
• Orbits of linear algebraic groups
• Log-linear models for frequency tables derived by indirect observation: Maximum likelihood equations
• An Iterative Procedure for Estimation in Contingency Tables
• Algebraic Statistics
• Biology
• Invariant Theory and Scaling Algorithms for Maximum Likelihood Estimation
• An elementary proof of the Hilbert-Mumford criterion
• Title not available (Why is that?)

Cited In (8)

• Bayesian Integrals on Toric Varieties
• Classical iterative proportional scaling of log-linear models with rational maximum likelihood estimator
• Invariant Theory and Scaling Algorithms for Maximum Likelihood Estimation
• Toric geometry of entropic regularization
• Maximum Likelihood Estimation from a Tropical and a Bernstein–Sato Perspective
• Maximum likelihood estimation of toric Fano varieties
• A bridge between invariant theory and maximum likelihood estimation
• Toric statistical models: parametric and binomial representations

This page was built for publication: Toric invariant theory for maximum likelihood estimation in log-linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2076293)