Fibers of multi-way contingency tables given conditionals: relation to marginals, cell bounds and Markov bases

DOI10.1007/S10463-014-0471-ZMaRDI QIDQ2355167zbMATH OpenOpenAlexFDO

Authors Xiaotian Zhu, Sonja Petrović, Aleksandra Slavković

Publication date 21 July 2015

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

Full work available at URL https://arxiv.org/abs/1401.1397

zbMATH Keywords

conditional tablescontingency tablesDiophantine equationsdirected acyclic graphsdisclosure limitationmarginal tablesMarkov basesoptimization for cell entries

Mathematics Subject Classification ID

Gröbner bases; other bases for ideals and modules (e.g., Janet and border bases) (13P10) Computational methods for problems pertaining to statistics (62-08) Contingency tables (62H17)

Abstract: A reference set, or a fiber, of a contingency table is the space of all realizations of the table under a given set of constraints such as marginal totals. Understanding the geometry of this space is a key problem in algebraic statistics, important for conducting exact conditional inference, calculating cell bounds, imputing missing cell values, and assessing the risk of disclosure of sensitive information. Motivated primarily by disclosure limitation problems where constraints can come from summary statistics other than the margins, in this paper we study the space $mathcal{F}_T$ of all possible multi-way contingency tables for a given sample size and set of observed conditional frequencies. We show that this space can be decomposed according to different possible marginals, which, in turn, are encoded by the solution set of a linear Diophantine equation. We characterize the difference between two fibers: $mathcal{F}_T$ and the space of tables for a given set of corresponding marginal totals. In particular, we solve a generalization of an open problem posed by Dobra et al. (2008). Our decomposition of $mathcal{F}_T$ has two important consequences: (1) we derive new cell bounds, some including connections to Directed Acyclic Graphs, and (2) we describe a structure for the Markov bases for the space $mathcal{F}_T$ that leads to a simplified calculation of Markov bases in this particular setting.

Recommendations

• Open problems on connectivity of fibers with positive margins in multi-dimensional contingency tables
• Markov bases of three-way tables are arbitrarily complicated
• Markov bases and subbases for bounded contingency tables
• Contingency tables of network type: models, Markov basis and applications
• Markov chains on the reference set of contingency tables with upper bounds

Cites work

• A Polynomial Time Algorithm for Counting Integral Points in Polyhedra When the Dimension is Fixed
• Algebraic algorithms for sampling from conditional distributions
• Algebraic statistics and contingency table problems: log-linear models, likelihood estimatio, and disclosure limitation
• An approximation algorithm for counting contingency tables
• Ancestral graph Markov models.
• Bounds for cell entries in contingency tables given marginal totals and decomposable graphs
• Computing generating sets of lattice ideals and Markov bases of lattices
• Conditional specification of statistical models.
• Data augmentation in multi-way contingency tables with fixed marginal totals
• Discrete Multivariate Analysis Theory and Practice
• Effective lattice point counting in rational convex polytopes
• Gröbner bases of lattices, corner polyhedra, and integer programming
• scientific article; zbMATH DE number 1817585 (Why is no real title available?)
• scientific article; zbMATH DE number 4211299 (Why is no real title available?)
• scientific article; zbMATH DE number 1134987 (Why is no real title available?)
• Lectures on algebraic statistics
• Markov bases for decomposable graphical models
• Markov bases of three-way tables are arbitrarily complicated
• Minimal Basis for a Connected Markov Chain over 3 x 3 x K Contingency Tables with Fixed Two-Dimensional Marginals
• Minimal invariant Markov basis for sampling contingency tables with fixed marginals
• On Systems of Linear Diophantine Equations
• On the number of Solutions of a Diophantine equation of Frobenius
• Relations among conditional probabilities
• Sequential importance sampling for multiway tables
• Simple Explicit Formula for Counting Lattice Points of Polyhedra
• Synthetic two-way contingency tables that preserve conditional frequencies
• The space of compatible full conditionals is a unimodular toric variety
• Using the Blankinship algorithm to find the general solution of a linear diophantine equation

Cited in

(8)

• On connectivity of fibers with positive marginals in multiple logistic regression
• Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs
• Markov bases of three-way tables are arbitrarily complicated
• A survey of discrete methods in (algebraic) statistics for networks
• Obtaining cell counts for contingency tables from rounded conditional frequencies
• New classes of degree sequences with fast mixing swap Markov chain sampling
• Generalized Fréchet Bounds for Cell Entries in Multidimensional Contingency Tables
• Contingency tables of network type: models, Markov basis and applications

Describes a project that uses

Uses Software

• 4ti2
• R
• LattE

This page was built for publication: Fibers of multi-way contingency tables given conditionals: relation to marginals, cell bounds and Markov bases

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2355167)