Maximum likelihood degree of variance component models

DOI10.1214/12-EJS702MaRDI QIDQ1950847zbMATH OpenOpenAlexWikidataFDO

Authors Elizabeth Gross, Mathias Drton, Sonja Petrović

Publication date 28 May 2013

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1111.3308, https://projecteuclid.org/euclid.ejs/1338469232

zbMATH Keywords

maximum likelihoodrestricted maximum likelihoodlinear mixed modelanalysis of variancevariance component

Mathematics Subject Classification ID

Point estimation (62F10) Analysis of variance and covariance (ANOVA) (62J10)

Abstract: Most statistical software packages implement numerical strategies for computation of maximum likelihood estimates in random effects models. Little is known, however, about the algebraic complexity of this problem. For the one-way layout with random effects and unbalanced group sizes, we give formulas for the algebraic degree of the likelihood equations as well as the equations for restricted maximum likelihood estimation. In particular, the latter approach is shown to be algebraically less complex. The formulas are obtained by studying a univariate rational equation whose solutions correspond to the solutions of the likelihood equations. Applying techniques from computational algebra, we also show that balanced two-way layouts with or without interaction have likelihood equations of degree four. Our work suggests that algebraic methods allow one to reliably find global optima of likelihood functions of linear mixed models with a small number of variance components.

Recommendations

• The maximum likelihood degree
• Solving the likelihood equations
• Maximum likelihood estimation in mixed normal models with two variance Components
• On Computing Maximum-Likelihood Estimates of the Unbalanced Two-Way Random-Effects Model
• Maximum likelihood degree of the two-dimensional linear Gaussian covariance model

Cites work

• scientific article; zbMATH DE number 4015954 (Why is no real title available?)
• scientific article; zbMATH DE number 47310 (Why is no real title available?)
• scientific article; zbMATH DE number 2154174 (Why is no real title available?)
• scientific article; zbMATH DE number 2162003 (Why is no real title available?)
• Composite binary losses
• Counting and Locating the Solutions of Polynomial Systems of Maximum Likelihood Equations, II: The Behrens-Fisher Problem
• Extending the linear model with R. Generalized linear, mixed effects and nonparametric regression models.
• Lectures on algebraic statistics
• Linear and generalized linear mixed models and their applications.
• Monte Carlo Comparison of ANOVA, MIVQUE, REML, and ML Estimators of Variance Components
• Open Problems in Algebraic Statistics
• Solving the likelihood equations
• The maximum likelihood degree
• Variance components.

Cited in

(8)

• Restricted likelihood representation for multivariate heterogeneous random effects models
• Selecting an augmented random effects model
• A probabilistic algorithm for computing data-discriminants of likelihood equations
• A survey of discrete methods in (algebraic) statistics for networks
• The Maximum Likelihood Degree of Sparse Polynomial Systems
• Linear optimization on varieties and Chern-Mather classes
• Maximum likelihood degree of the two-dimensional linear Gaussian covariance model
• The maximum likelihood degree of mixtures of independence models

Describes a project that uses

Uses Software

• faraway

This page was built for publication: Maximum likelihood degree of variance component models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1950847)