Invariant theory and scaling algorithms for maximum likelihood estimation
From MaRDI portal
Publication:6337697
DOI10.1137/20M1328932arXiv2003.13662MaRDI QIDQ6337697FDOQ6337697
Authors: Kathlén Kohn, Philipp Reichenbach, Anna Leah Seigal, Carlos Enrique Améndola Cerón 
Publication date: 30 March 2020
Abstract: We uncover connections between maximum likelihood estimation in statistics and norm minimization over a group orbit in invariant theory. We focus on Gaussian transformation families, which include matrix normal models and Gaussian graphical models given by transitive directed acyclic graphs. We use stability under group actions to characterize boundedness of the likelihood, and existence and uniqueness of the maximum likelihood estimate. Our approach reveals promising consequences of the interplay between invariant theory and statistics. In particular, existing scaling algorithms from statistics can be used in invariant theory, and vice versa.
Point estimation (62F10) Probabilistic graphical models (62H22) Algebraic statistics (62R01) Real algebraic sets (14P05) Geometric invariant theory (14L24) Applications of linear algebraic groups to the sciences (20G45)
This page was built for publication: Invariant theory and scaling algorithms for maximum likelihood estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6337697)
