The Annals of Statistics

Markovian acyclic directed mixed graphs for discrete data

Robin J. Evans and Thomas S. Richardson

Full-text: Open access


Acyclic directed mixed graphs (ADMGs) are graphs that contain directed ($\rightarrow$) and bidirected ($\leftrightarrow$) edges, subject to the constraint that there are no cycles of directed edges. Such graphs may be used to represent the conditional independence structure induced by a DAG model containing hidden variables on its observed margin. The Markovian model associated with an ADMG is simply the set of distributions obeying the global Markov property, given via a simple path criterion (m-separation). We first present a factorization criterion characterizing the Markovian model that generalizes the well-known recursive factorization for DAGs. For the case of finite discrete random variables, we also provide a parameterization of the model in terms of simple conditional probabilities, and characterize its variation dependence. We show that the induced models are smooth. Consequently, Markovian ADMG models for discrete variables are curved exponential families of distributions.

Article information

Ann. Statist. Volume 42, Number 4 (2014), 1452-1482.

First available in Project Euclid: 7 August 2014

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62M45: Neural nets and related approaches

Acyclic directed mixed graph curved exponential family conditional independence graphical model m-separation parameterization


Evans, Robin J.; Richardson, Thomas S. Markovian acyclic directed mixed graphs for discrete data. Ann. Statist. 42 (2014), no. 4, 1452--1482. doi:10.1214/14-AOS1206.

Export citation


  • Andersson, S. A., Madigan, D. and Perlman, M. D. (2001). Alternative Markov properties for chain graphs. Scand. J. Stat. 28 33–85.
  • Bartolucci, F., Colombi, R. and Forcina, A. (2007). An extended class of marginal link functions for modelling contingency tables by equality and inequality constraints. Statist. Sinica 17 691–711.
  • Bergsma, W. P. and Rudas, T. (2002). Marginal models for categorical data. Ann. Statist. 30 140–159.
  • Dawid, A. P. (1979). Conditional independence in statistical theory. J. Roy. Statist. Soc. Ser. B 41 1–31.
  • Dawid, A. P. and Didelez, V. (2010). Identifying the consequences of dynamic treatment strategies: A decision-theoretic overview. Stat. Surv. 4 184–231.
  • Drton, M. (2009). Discrete chain graph models. Bernoulli 15 736–753.
  • Drton, M. and Xiao, H. (2010). Smoothness of Gaussian conditional independence models. In Algebraic Methods in Statistics and Probability II. Contemp. Math. 516 155–177. Amer. Math. Soc., Providence, RI.
  • Evans, R. J. and Richardson, T. S. (2010). Maximum likelihood fitting of acyclic directed mixed graphs to binary data. In Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence 177–184. AUAI Press, Corvallis, OR.
  • Evans, R. J. and Richardson, T. S. (2013). Marginal log-linear parameters for graphical Markov models. J. R. Stat. Soc. Ser. B Stat. Methodol. 75 743–768.
  • Hammersley, J. M. and Clifford, P. (1971). Markov fields on finite graphs and lattices. Unpublished manuscript.
  • Huang, Y. and Valtorta, M. (2006). Pearl’s calculus of interventions is complete. In Proceedings of the 22nd Conference On Uncertainty in Artificial Intelligence. AUAI Press, Arlington, VA.
  • Kass, R. E. and Vos, P. W. (1997). Geometrical Foundations of Asymptotic Inference. Wiley, New York.
  • Lauritzen, S. L. (1996). Graphical Models. Oxford Univ. Press, New York.
  • Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Mateo, CA.
  • Pearl, J. (1995). Causal diagrams for empirical research. Biometrika 82 669–710.
  • Pearl, J. (2009). Causality, 2nd ed. Cambridge Univ. Press, Cambridge.
  • Pearl, J. and Robins, J. (1995). Probabilistic evaluation of sequential plans from causal models with hidden variables. In Uncertainty in Artificial Intelligence (Montreal, PQ, 1995) 444–453. Morgan Kaufmann, San Francisco, CA.
  • Richardson, T. (2003). Markov properties for acyclic directed mixed graphs. Scand. J. Stat. 30 145–157.
  • Richardson, T. S. (2009). A factorization criterion for acyclic directed mixed graphs. In Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence 462–470. AUAI Press, Arlington, VA.
  • Richardson, T. and Spirtes, P. (2002). Ancestral graph Markov models. Ann. Statist. 30 962–1030.
  • Robins, J. M. and Richardson, T. S. (2011). Alternative graphical causal models and the identification of direct effects. In Causality and Psychopathology: Finding the Determinants of Disorders and Their Cures (P. Shrout, K. Keyes and K. Ornstein, eds.) 6 1–52. Oxford Univ. Press, London.
  • Sadeghi, K. (2013). Stable mixed graphs. Bernoulli 19 2330–2358.
  • Sadeghi, K. and Lauritzen, S. (2014). Markov properties for mixed graphs. Bernoulli 20 676–696.
  • Shpitser, I. and Pearl, J. (2006a). Identification of joint interventional distributions in recursive semi-Markovian causal models. In Proceedings of the 21st National Conference on Artificial Intelligence. AAAI Press, Menlo Park, CA.
  • Shpitser, I. and Pearl, J. (2006b). Identification of conditional interventional distributions. In Proceedings of the 22nd Conference on Uncertainty in Artificial Intelligence 437–444. AUAI Press, Arlington, VA.
  • Silva, R. and Ghahramani, Z. (2009). The hidden life of latent variables: Bayesian learning with mixed graph models. J. Mach. Learn. Res. 10 1187–1238.
  • Spirtes, P., Glymour, C. and Scheines, R. (1993). Causation, Prediction, and Search. Lecture Notes in Statistics 81. Springer, New York.
  • Tian, J. and Pearl, J. (2002). A general identification condition for causal effects. In Proceedings of the 18th National Conference on Artificial Intelligence. AAAI Press, Menlo Park.
  • Wermuth, N. (2011). Probability distributions with summary graph structure. Bernoulli 17 845–879.