Markov bases for decomposable graphical models
Open Access
- 1 December 2003
- journal article
- Published by Bernoulli Society for Mathematical Statistics and Probability in Bernoulli
- Vol. 9 (6) , 1093-1108
- https://doi.org/10.3150/bj/1072215202
Abstract
In this paper we show that primitive data swaps or moves are the only moves that have to be included in a Markov basis that links all the contingency tables having a set of fixed marginals when this set of marginals induce a decomposable independence graph. We give formulas that fully identify such Markov bases and show how to use these formulas to dynamically generate random moves.Keywords
This publication has 14 references indexed in Scilit:
- Bounds for cell entries in contingency tables given marginal totals and decomposable graphsProceedings of the National Academy of Sciences, 2000
- The Diaconis-Sturmfels Algorithm and Rules of SuccessionBernoulli, 1998
- Algebraic algorithms for sampling from conditional distributionsThe Annals of Statistics, 1998
- Bayesian Graphical Models for Discrete DataInternational Statistical Review, 1995
- Rectangular Arrays with Fixed MarginsPublished by Springer Nature ,1995
- The exact analysis of contingency tables in medical researchStatistical Methods in Medical Research, 1994
- An Introduction to Chordal Graphs and Clique TreesPublished by Springer Nature ,1993
- A Survey of Exact Inference for Contingency TablesStatistical Science, 1992
- Conditions for the existence of solutions of the three-dimensional planar transportation problemDiscrete Applied Mathematics, 1986
- Testing for Independence in a Two-Way Table: New Interpretations of the Chi-Square StatisticThe Annals of Statistics, 1985