A divide-and-conquer algorithm for generating Markov bases of multi-way tables
The paper focuses on sets of \(k\)-way tables of counts that are induced by fixing several lower dimensional marginals. The authors demonstrate how using the graphical structure of the mentioned class of tables can significantly reduce the amount of computations necessary for generating a Markov basic. They describe a relevant ``divide-and-conquer procedure based on decomposition of the independence graph induced by fixed marginals, study properties of decomposable and reducible graphs and propose a divide-and-conquer algorithm. They also discuss the practical importance of using such method in conjunction with other algorithms for generating Markov bases.
- Minimal Basis for a Connected Markov Chain over 3 x 3 x K Contingency Tables with Fixed Two-Dimensional Marginals
- Markov bases for decomposable graphical models
- Open problems on connectivity of fibers with positive margins in multi-dimensional contingency tables
- Minimal and minimal invariant Markov bases of decomposable models for contingency tables
- Markov bases of binary graph models
- A survey of exact inference for contingency tables. With comments and a rejoinder by the author
- Algebraic algorithms for sampling from conditional distributions
- Algebraic unimodular counting
- Bayesian Graphical Models for Discrete Data
- Bounds for cell entries in contingency tables given marginal totals and decomposable graphs
- Conditions for the existence of solutions of the three-dimensional planar transportation problem
- Data-swapping: A technique for disclosure control
- Decomposition by clique separators
- Elements of statistical disclosure control
- Gröbner bases and polyhedral geometry of reducible and cyclic models.
- scientific article; zbMATH DE number 4211299 (Why is no real title available?)
- scientific article; zbMATH DE number 108068 (Why is no real title available?)
- scientific article; zbMATH DE number 3519741 (Why is no real title available?)
- scientific article; zbMATH DE number 554762 (Why is no real title available?)
- scientific article; zbMATH DE number 1134987 (Why is no real title available?)
- scientific article; zbMATH DE number 795108 (Why is no real title available?)
- Markov bases for decomposable graphical models
- Optimal decomposition by clique separators
- Testing for independence in a two-way table: New interpretations of the chi-square statistic
- The Diaconis-Sturmfels algorithm and rules of succession
- Markov bases for two-way subtable sum problems
- Hybrid schemes for exact conditional inference in discrete exponential families
- Gröbner bases and polyhedral geometry of reducible and cyclic models.
- Some characterizations of minimal Markov basis for sampling from discrete conditional distribu\-tions
- Gaussian graphical models with toric vanishing ideals
- Multigraded commutative algebra of graph decompositions
- Conditions for swappability of records in a microdata set when some marginals are fixed
- Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs
- Toric fiber products
- Markov bases and subbases for bounded contingency tables
- Combinatorial degree bound for toric ideals of hypergraphs
- A survey of discrete methods in (algebraic) statistics for networks
- Connecting tables with zero-one entries by a subset of a Markov basis
- Markov bases for typical block effect models of two-way contingency tables
- Power analysis of independence testing for three-way contingency tables of small sizes
- Minimal and minimal invariant Markov bases of decomposable models for contingency tables
- Hierarchical subspace models for contingency tables
- Conditional inference given partial information in contingency tables using Markov bases
- Markov Bases: A 25 Year Update
- Equivariant lattice bases
- New directions in algebraic statistics: three challenges from 2023
- Perturbation method for determining the group of invariance of hierarchical models
- A finiteness theorem for Markov bases of hierarchical models
- A Markov basis for conditional test of common diagonal effect in quasi-independence model for square contingency tables
This page was built for publication: A divide-and-conquer algorithm for generating Markov bases of multi-way tables
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1775965)
