Abstract
This paper deals with multivariate regression chain graphs (MVR CGs), which were introduced by Cox and Wermuth in the nineties to represent linear causal models with correlated errors. We consider the PC-like algorithm for structure learning of MVR CGs, a constraint-based method proposed by Sonntag and Peña in 2012. We show that the PC-like algorithm is order-dependent, because the output can depend on the order in which the variables are given. This order-dependence is a minor issue in low-dimensional settings. However, it can be very pronounced in high-dimensional settings, where it can lead to highly variable results. We propose two modifications of the PC-like algorithm that remove part or all of this order-dependence. Simulations under a variety of settings demonstrate the competitive performance of our algorithms in comparison with the original PC-like algorithm in low-dimensional settings and improved performance in high-dimensional settings.
Supported by AFRL and DARPA (FA8750-16-2-0042).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Andersson, S.A., Madigan, D., Perlman, M.D.: An alternative Markov property for chain graphs. In: Proceedings of UAI Conference, pp. 40–48 (1996)
Colombo, D., Maathuis, M.H.: Order-independent constraint-based causal structure learning. J. Mach. Learn. Res. 15(1), 3741–3782 (2014)
Cox, D.R., Wermuth, N.: Linear dependencies represented by chain graphs. Stat. Sci. 8(3), 204–218 (1993)
Drton, M.: Discrete chain graph models. Bernoulli 15(3), 736–753 (2009)
Frydenberg, M.: The chain graph Markov property. Scand. J. Stat. 17(4), 333–353 (1990)
Golumbic, M.: Algorithmic Graph Theory and Perfect Graphs. Academic Press, New York (1980)
Javidian, M.A., Valtorta, M.: On the properties of MVR chain graphs. In: Workshop Proceedings of PGM Conference, pp. 13–24 (2018)
Kalisch, M., Bühlmann, P.: Estimating high-dimensional directed acyclic graphs with the PC-algorithm. J. Mach. Learn. Res. 8, 613–636 (2007)
Lauritzen, S., Wermuth, N.: Graphical models for associations between variables, some of which are qualitative and some quantitative. Ann. Stat. 17(1), 31–57 (1989)
Ma, Z., Xie, X., Geng, Z.: Structural learning of chain graphs via decomposition. J. Mach. Learn. Res. 9, 2847–2880 (2008)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA (1988)
Pearl, J.: Causality: Models, Reasoning, and Inference. Cambridge University Press, Cambridge (2009)
Peña, J.M.: Alternative Markov and causal properties for acyclic directed mixed graphs. In: Proceedings of UAI Conference, pp. 577–586 (2016)
Ramsey, J., Spirtes, P., Zhang, J.: Adjacency-faithfulness and conservative causal inference. In: Proceedings of UAI Conference, pp. 401–408 (2006)
Richardson, T.S.: Markov properties for acyclic directed mixed graphs. Scand. J. Stat. 30(1), 145–157 (2003)
Sonntag, D., Peña, J.M.: Learning multivariate regression chain graphs under faithfulness. In: Proceedings of PGM Workshop, pp. 299–306 (2012)
Sonntag, D., Peña, J.M.: Chain graph interpretations and their relations revisited. Int. J. Approx. Reason. 58, 39–56 (2015)
Sonntag, D., Peña, J.M., Gómez-Olmedo, M.: Approximate counting of graphical models via MCMC revisited. Int. J. Intell. Syst. 30(3), 384–420 (2015)
Spirtes, P., Glymour, C., Scheines, R.: Causation, Prediction and Search, 2nd edn. MIT Press, Cambridge (2000)
Tsamardinos, I., Brown, L.E., Aliferis, C.F.: The max-min hill-climbing Bayesian network structure learning algorithm. Mach. Learn. 65(1), 31–78 (2006)
Wermuth, N., Sadeghi, K.: Sequences of regressions and their independences. Test 21, 215–252 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Javidian, M.A., Valtorta, M., Jamshidi, P. (2019). Order-Independent Structure Learning of Multivariate Regression Chain Graphs. In: Ben Amor, N., Quost, B., Theobald, M. (eds) Scalable Uncertainty Management. SUM 2019. Lecture Notes in Computer Science(), vol 11940. Springer, Cham. https://doi.org/10.1007/978-3-030-35514-2_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-35514-2_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-35513-5
Online ISBN: 978-3-030-35514-2
eBook Packages: Computer ScienceComputer Science (R0)