Marginal likelihood and model selection for Gaussian latent tree and forest models

DOI10.3150/15-BEJ775zbMATH Open1381.62224arXiv1412.8285OpenAlexW2962774697WikidataQ57566353 ScholiaQ57566353MaRDI QIDQ520697FDOQ520697

Luca Weihs, Mathias Drton, Shaowei Lin, Piotr Zwiernik

Publication date: 5 April 2017

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Gaussian latent tree models, or more generally, Gaussian latent forest models have Fisher-information matrices that become singular along interesting submodels, namely, models that correspond to subforests. For these singularities, we compute the real log-canonical thresholds (also known as stochastic complexities or learning coefficients) that quantify the large-sample behavior of the marginal likelihood in Bayesian inference. This provides the information needed for a recently introduced generalization of the Bayesian information criterion. Our mathematical developments treat the general setting of Laplace integrals whose phase functions are sums of squared differences between monomials and constants. We clarify how in this case real log-canonical thresholds can be computed using polyhedral geometry, and we show how to apply the general theory to the Laplace integrals associated with Gaussian latent tree and forest models. In simulations and a data example, we demonstrate how the mathematical knowledge can be applied in model selection.

Full work available at URL: https://arxiv.org/abs/1412.8285

Recommendations

• The correlation space of Gaussian latent tree models and model selection without fitting
• scientific article; zbMATH DE number 3956179
• scientific article
• Posterior concentration for Bayesian regression trees and forests
• Latent tree models
• On the Bayesian treed multivariate Gaussian process with linear model of coregionalization
• scientific article; zbMATH DE number 6253886
• scientific article; zbMATH DE number 1799346
• Model selection and estimation in the Gaussian graphical model

zbMATH Keywords

multivariate normal distributionGaussian graphical modelmarginal likelihoodalgebraic statisticslatent tree modelssingular learning theory

Mathematics Subject Classification ID

Bayesian inference (62F15) Multivariate analysis (62H99) Parametric inference (62F99)

Cited In (3)

• The exact asymptotic form of Bayesian generalization error in latent Dirichlet allocation
• Consideration on the learning efficiency of multiple-layered neural networks with linear units
• Variational approximation error in non-negative matrix factorization

This page was built for publication: Marginal likelihood and model selection for Gaussian latent tree and forest models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q520697)