A structural Markov property for decomposable graph laws that allows control of clique intersections

Peter Green, Alun Thomas

Research output: Contribution to journalArticle (Academic Journal)peer-review

201 Downloads (Pure)

Abstract

We present a new kind of structural Markov property for probabilistic laws on decomposable graphs, which allows the explicit control of interactions between cliques and so is capable of encoding some interesting structure. We prove the equivalence of this property to an exponential family assumption, and discuss identifiability, modelling, inferential and computational implications.
Original languageEnglish
Pages (from-to)19-29
Number of pages11
JournalBiometrika
Volume105
Issue number1
Early online date11 Dec 2017
DOIs
Publication statusPublished - 1 Mar 2018

Keywords

  • Conditional independence
  • Graphical model
  • Hub model
  • Markov random field
  • Model determination
  • Random graph

Fingerprint Dive into the research topics of 'A structural Markov property for decomposable graph laws that allows control of clique intersections'. Together they form a unique fingerprint.

Cite this