Faithfulness and learning hypergraphs from discrete distributions

DOI10.1016/J.CSDA.2015.01.017zbMATH Open1468.62103arXiv1404.6617OpenAlexW2031537979MaRDI QIDQ1663308FDOQ1663308

Anna Klimova, Tamás Rudas, Caroline Uhler

Publication date: 21 August 2018

Published in: Computational Statistics and Data Analysis (Search for Journal in Brave)

Abstract: The concepts of faithfulness and strong-faithfulness are important for statistical learning of graphical models. Graphs are not sufficient for describing the association structure of a discrete distribution. Hypergraphs representing hierarchical log-linear models are considered instead, and the concept of parametric (strong-) faithfulness with respect to a hypergraph is introduced. Strong-faithfulness ensures the existence of uniformly consistent parameter estimators and enables building uniformly consistent procedures for a hypergraph search. The strength of association in a discrete distribution can be quantified with various measures, leading to different concepts of strong-faithfulness. Lower and upper bounds for the proportions of distributions that do not satisfy strong-faithfulness are computed for different parameterizations and measures of association.

Full work available at URL: https://arxiv.org/abs/1404.6617

Recommendations

• A note on faithfulness and total positivity
• Faithfulness of probability distributions and graphs
• Faithfulness in chain graphs: the discrete case
• Geometry of the faithfulness assumption in causal inference
• Learning unfaithful \(K\)-separable Gaussian graphical models

zbMATH Keywords

contingency tableshypergraphsdirected acyclic graphshierarchical log-linear models(strong-)faithfulness

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Probabilistic graphical models (62H22)

Cites Work

• Introduction to Graphical Modelling
• Title not available (Why is that?)
• Title not available (Why is that?)
• Uniform consistency in causal inference
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Measures of Association for Cross Classifications
• Geometry of the faithfulness assumption in causal inference
• Marginal models for categorical data
• Causation, prediction, and search. With additional material by David Heckerman, Christopher Meek, Gregory F. Cooper and Thomas Richardson.
• Analogies between Multiplicative Models in Contingency Tables and Covariance Selection
• An algorithm for reading dependencies from the minimal undirected independence map of a graphoid that satisfies weak transitivity
• A note on adding and deleting edges in hierarchical log-linear models
• Parameterization and estimation of path models for categorical data

Cited In (2)

• Implications of faithfulness in graphical models
• Faithfulness in chain graphs: the discrete case

Uses Software

• pcalg
• ElemStatLearn
• TETRAD
• MIM

This page was built for publication: Faithfulness and learning hypergraphs from discrete distributions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1663308)