Differentiating the pseudo determinant

DOI10.1016/J.LAA.2018.03.018MaRDI QIDQ1748089zbMATH OpenOpenAlexWikidataFDO

Authors Andrew J. Holbrook

Publication date 2 May 2018

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

Full work available at URL https://arxiv.org/abs/1802.04878

zbMATH Keywords

maximum likelihoodpseudo inversedegenerate Gaussianpseudo determinantsingular covariance

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Determinants, permanents, traces, other special matrix functions (15A15)

Abstract: A class of derivatives is defined for the pseudo determinant $Det(A)$ of a Hermitian matrix $A$. This class is shown to be non-empty and to have a unique, canonical member $mathbfablaDet(A)=Det(A)A^{+}$, where $A^{+}$ is the Moore-Penrose pseudo inverse. The classic identity for the gradient of the determinant is thus reproduced. Examples are provided, including the maximum likelihood problem for the rank-deficient covariance matrix of the degenerate multivariate Gaussian distribution.

Recommendations

• Pseudo-inverse multivariate/matrix-variate distributions
• Cauchy-Binet for pseudo-determinants
• The Drazin inverse as a gradient
• Higher order derivatives and perturbation bounds for determinants
• scientific article; zbMATH DE number 1537552

Cites work

• scientific article; zbMATH DE number 192992 (Why is no real title available?)
• scientific article; zbMATH DE number 2038320 (Why is no real title available?)
• Cauchy-Binet for pseudo-determinants
• Geodesic Lagrangian Monte Carlo over the space of positive definite matrices: with application to Bayesian spectral density estimation
• Maximum-likelihood estimation of the parameters of a multivariate normal distribution
• The Differentiation of Pseudo-Inverses and Nonlinear Least Squares Problems Whose Variables Separate
• Three results in connection with inverse matrices

Cited in

(4)

• Total positivity in multivariate extremes
• Generating MCMC proposals by randomly rotating the regular simplex
• Optimal designs for dose-escalation trials and individual allocations in cohorts
• A singular woodbury and pseudo-determinant matrix identities and application to Gaussian process regression

This page was built for publication: Differentiating the pseudo determinant

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1748089)