Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs

DOI10.1007/S10463-016-0560-2zbMATH Open1400.62124arXiv1401.4896OpenAlexW2962856988MaRDI QIDQ2397339FDOQ2397339

Authors: Elizabeth Gross, Sonja Petrović, Despina Stasi 

Publication date: 22 May 2017

Published in: Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Abstract: Social networks and other large sparse data sets pose significant challenges for statistical inference, as many standard statistical methods for testing model fit are not applicable in such settings. Algebraic statistics offers a theoretically justified approach to goodness-of-fit testing that relies on the theory of Markov bases and is intimately connected with the geometry of the model as described by its fibers. Most current practices require the computation of the entire basis, which is infeasible in many practical settings. We present a dynamic approach to explore the fiber of a model, which bypasses this issue, and is based on the combinatorics of hypergraphs arising from the toric algebra structure of log-linear models. We demonstrate the approach on the Holland-Leinhardt $p_1$ model for random directed graphs that allows for reciprocated edges.

Full work available at URL: https://arxiv.org/abs/1401.4896

Recommendations

• A survey of discrete methods in (algebraic) statistics for networks
• Algebraic statistics for a directed random graph model with reciprocation
• Exact tests for singular network data
• scientific article; zbMATH DE number 434919
• Goodness of Fit of Social Network Models

zbMATH Keywords

contingency tablesocial networksrandom graphalgebraic statisticsnetwork modeltoric idealhypergraphMarkov basissampling algorithm

Mathematics Subject Classification ID

Contingency tables (62H17) Random graphs (graph-theoretic aspects) (05C80) Social networks; opinion dynamics (91D30) Applications of commutative algebra (e.g., to statistics, control theory, optimization, etc.) (13P25)

Cites Work

• A survey of statistical network models
• Goodness of Fit of Social Network Models
• An Exponential Family of Probability Distributions for Directed Graphs
• Title not available (Why is that?)
• Title not available (Why is that?)
• Lectures on algebraic statistics
• Title not available (Why is that?)
• Sequential importance sampling for multiway tables
• Random graphs with a given degree sequence
• Algebraic algorithms for sampling from conditional distributions
• Monomial algebras
• Markov bases for decomposable graphical models
• Fibers of multi-way contingency tables given conditionals: relation to marginals, cell bounds and Markov bases
• Graver basis for an undirected graph and its application to testing the beta model of random graphs
• Connecting tables with zero-one entries by a subset of a Markov basis
• Algebraic statistics for a directed random graph model with reciprocation
• Sampling large tables with constraints
• Markov bases and subbases for bounded contingency tables
• Commutative algebra of statistical ranking
• Markov bases of binary graph models of \(K_{4}\)-minor free graphs
• Markov degree of the Birkhoff model
• Minimal Basis for a Connected Markov Chain over 3 x 3 x K Contingency Tables with Fixed Two-Dimensional Marginals
• Markov chain Monte Carlo exact tests for incomplete two-way contingency tables
• Markov bases of binary graph models
• Markov bases for two-way subtable sum problems
• A divide-and-conquer algorithm for generating Markov bases of multi-way tables
• Algebraic statistics and contingency table problems: log-linear models, likelihood estimatio, and disclosure limitation
• Minimal and minimal invariant Markov bases of decomposable models for contingency tables
• A Markov basis for conditional test of common diagonal effect in quasi-independence model for square contingency tables
• The three-state toric homogeneous Markov chain model has Markov degree two
• Toric algebra of hypergraphs
• Combinatorial degree bound for toric ideals of hypergraphs
• Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs
• Markov degree of the three-state toric homogeneous Markov chain model

Cited In (7)

• Markov Bases: A 25 Year Update
• Blow-up algebras, determinantal ideals, and Dedekind-Mertens-like formulas
• A survey of discrete methods in (algebraic) statistics for networks
• New classes of degree sequences with fast mixing swap Markov chain sampling
• Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs
• What Are Higher-Order Networks?
• Gibbs regression and a test for goodness-of-fit

Uses Software

• R
• Pajek datasets
• 4ti2
• igraph

This page was built for publication: Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2397339)