A higher-order LQ decomposition for separable covariance models

DOI10.1016/J.LAA.2016.04.033zbMATH Open1360.15030arXiv1410.1094OpenAlexW2161436879MaRDI QIDQ149116FDOQ149116

Authors: David Gerard, Peter D. Hoff 

Publication date: September 2016

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

Abstract: We develop a higher order generalization of the LQ decomposition and show that this decomposition plays an important role in likelihood-based estimation and testing for separable, or Kronecker structured, covariance models, such as the multilinear normal model. This role is analogous to that of the LQ decomposition in likelihood inference for the multivariate normal model. Additionally, this higher order LQ decomposition can be used to construct an alternative version of the popular higher order singular value decomposition for tensor-valued data. We also develop a novel generalization of the polar decomposition to tensor-valued data.

Full work available at URL: https://arxiv.org/abs/1410.1094

Recommendations

• More on the Kronecker structured covariance matrix
• The likelihood ratio test for a separable covariance matrix
• A Multilinear Singular Value Decomposition
• Estimating Higher-Order Moments Using Symmetric Tensor Decomposition
• A new tensor decomposition

zbMATH Keywords

tensorsingular value decompositionlikelihood ratio testLQ decompositionpolar decompositionTucker decomposition

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Hypothesis testing in multivariate analysis (62H15) Multilinear algebra, tensor calculus (15A69) Numerical linear algebra (65F99)

Cites Work

• Maximum likelihood estimation for the tensor normal distribution: Algorithm, minimum sample size, and empirical bias and dispersion
• Equivariant minimax dominators of the MLE in the array normal model
• Convergence of a block coordinate descent method for nondifferentiable minimization
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Testing Statistical Hypotheses
• Title not available (Why is that?)
• Tensor Decompositions and Applications
• Some matrix-variate distribution theory: Notational considerations and a Bayesian application
• Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation
• The likelihood ratio test for a separable covariance matrix
• Separable covariance arrays via the Tucker product, with applications to multivariate relational data
• Tensor approximation and signal processing applications
• Array variate random variables with multiway Kronecker delta covariance matrix structure
• Hierarchical Singular Value Decomposition of Tensors
• Maximum-likelihood estimates and likelihood-ratio criteria for multivariate elliptically contoured distributions
• Title not available (Why is that?)
• A Multilinear Singular Value Decomposition
• On the Best Rank-1 and Rank-(R₁ ,R₂ ,. . .,R_N) Approximation of Higher-Order Tensors
• Geodesic Convexity and Covariance Estimation
• Equivariant and scale-free Tucker decomposition models
• Model Averaging and Dimension Selection for the Singular Value Decomposition
• Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
• Factorization strategies for third-order tensors
• The multilinear normal distribution: introduction and some basic properties

Cited In (3)

• Triangular decomposition of CP factors of a third-order tensor with application to solving nonlinear systems of equations
• On the largest multilinear singular values of higher-order tensors
• tensr

Uses Software

• R

This page was built for publication: A higher-order LQ decomposition for separable covariance models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q149116)