GEOMETRY OF MAXIMUM LIKELIHOOD ESTIMATION IN GAUSSIAN GRAPHICAL MODELS

被引：39

作者：

Uhler, Caroline ^{[1
]}

机构：

[1] IST Austria, A-3400 Klosterneuburg, Austria

来源：

ANNALS OF STATISTICS | 2012年 / 40卷 / 01期

关键词：

Gaussian graphical model; maximum likelihood estimation; matrix completion problems; duality; algebraic statistics; algebraic variety; number of observations; sufficient statistics; treewidth; elimination ideal; ML degree; bipartite graphs;

D O I：

10.1214/11-AOS957

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We study maximum likelihood estimation in Gaussian graphical models from a geometric point of view. An algebraic elimination criterion allows us to find exact lower bounds on the number of observations needed to ensure that the maximum likelihood estimator (MLE) exists with probability one. This is applied to bipartite graphs, grids and colored graphs. We also study the ML degree, and we present the first instance of a graph for which the MLE exists with probability one, even when the number of observations equals the treewidth.

引用

页码：238 / 261

页数：24

共 50 条

[31] Distributed Covariance Estimation in Gaussian Graphical Models
Wiesel, Ami
Hero, Alfred O., III
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2012, 60 (01) : 211 - 220
[32] Penalised maximum likelihood estimation for fractional Gaussian processes
Lieberman, O
[J]. BIOMETRIKA, 2001, 88 (03) : 888 - 894
[33] On some algorithms for estimation in Gaussian graphical models
Hojsgaard, S.
Lauritzen, S.
[J]. BIOMETRIKA, 2024,
[34] Covariance Estimation in Decomposable Gaussian Graphical Models
Wiesel, Ami
Eldar, Yonina C.
Hero, Alfred O., III
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2010, 58 (03) : 1482 - 1492
[35] FLEXIBLE COVARIANCE ESTIMATION IN GRAPHICAL GAUSSIAN MODELS
Rajaratnam, Bala
Massam, Helene
Carvalho, Carlos M.
[J]. ANNALS OF STATISTICS, 2008, 36 (06): : 2818 - 2849
[36] ESTIMATION OF MEANS IN GRAPHICAL GAUSSIAN MODELS WITH SYMMETRIES
Gehrmann, Helene
Lauritzen, Steffen L.
[J]. ANNALS OF STATISTICS, 2012, 40 (02): : 1061 - 1073
[37] Graphical methods for efficient likelihood inference in Gaussian covariance models
Drton, Mathias
Richardson, Thomas S.
[J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2008, 9 : 893 - 914
[38] DECOMPOSITION OF MAXIMUM-LIKELIHOOD IN MIXED GRAPHICAL INTERACTION MODELS
FRYDENBERG, M
LAURITZEN, SL
[J]. BIOMETRIKA, 1989, 76 (03) : 539 - 555
[39] High-Order Analysis of the Efficiency Gap for Maximum Likelihood Estimation in Nonlinear Gaussian Models
Yeredor, Arie
Weiss, Amir
Weiss, Anthony J.
[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2018, 66 (18) : 4782 - 4795
[40] Empirical Bayes hierarchical models for regularizing maximum likelihood estimation in the matrix Gaussian Procrustes problem
Theobald, Douglas L.
Wuttke, Deborah S.
[J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2006, 103 (49) : 18521 - 18527

← 1 2 3 4 5 →