Graphical Markov models for infinitely many variables
HTML articles powered by AMS MathViewer
- by David Montague and Bala Rajaratnam PDF
- Trans. Amer. Math. Soc. 370 (2018), 7557-7603 Request permission
Abstract:
Representing the conditional independences present in a multivariate random vector via graphs has found widespread use in applications, and such representations are popularly known as graphical models or Markov random fields. These models have many useful properties, but their fundamental attractive feature is their ability to reflect conditional independences between blocks of variables through graph separation, a consequence of the equivalence of the pairwise, local, and global Markov properties demonstrated by Pearl and Paz (1985). Modern-day applications often necessitate working with either an infinite collection of variables (such as in a spatial-temporal field) or approximating a large high-dimensional finite stochastic system with an infinite-dimensional system. However, it is unclear whether the conditional independences present in an infinite-dimensional random vector or stochastic process can still be represented by separation criteria in an infinite graph. In light of the advantages of using graphs as tools to represent stochastic relationships, we undertake in this paper a general study of infinite graphical models. First, we demonstrate that naïve extensions of the assumptions required for the finite case results do not yield equivalence of the Markov properties in the infinite-dimensional setting, thus calling for a more in-depth analysis. To this end, we proceed to derive general conditions which do allow representing the conditional independence in an infinite-dimensional random system by means of graphs, and our results render the result of Pearl and Paz as a special case of a more general phenomenon. We conclude by demonstrating the applicability of our theory through concrete examples of infinite-dimensional graphical models.References
- V. I. Bogachev, Measure theory. Vol. I, II, Springer-Verlag, Berlin, 2007. MR 2267655, DOI 10.1007/978-3-540-34514-5
- C. Borgs, J. T. Chayes, L. Lovász, V. T. Sós, and K. Vesztergombi, Convergent sequences of dense graphs. I. Subgraph frequencies, metric properties and testing, Adv. Math. 219 (2008), no. 6, 1801–1851. MR 2455626, DOI 10.1016/j.aim.2008.07.008
- C. Borgs, J. T. Chayes, L. Lovász, V. T. Sós, and K. Vesztergombi, Convergent sequences of dense graphs II. Multiway cuts and statistical physics, Ann. of Math. (2) 176 (2012), no. 1, 151–219. MR 2925382, DOI 10.4007/annals.2012.176.1.2
- A. P. Dawid, Conditional independence in statistical theory, J. Roy. Statist. Soc. Ser. B 41 (1979), no. 1, 1–31. MR 535541
- A. Philip Dawid, Conditional independence for statistical operations, Ann. Statist. 8 (1980), no. 3, 598–617. MR 568723
- R. L. Dobrušin, Description of a random field by means of conditional probabilities and conditions for its regularity, Teor. Verojatnost. i Primenen 13 (1968), 201–229 (Russian, with English summary). MR 0231434
- J. L. Doob, Stochastic processes, John Wiley & Sons, Inc., New York; Chapman & Hall, Ltd., London, 1953. MR 0058896
- Hans-Otto Georgii, Gibbs measures and phase transitions, De Gruyter Studies in Mathematics, vol. 9, Walter de Gruyter & Co., Berlin, 1988. MR 956646, DOI 10.1515/9783110850147
- Robert M. Gray, Toeplitz and circulant matrices: A review, Foundations and Trends in Communications and Information Theory 2 (2006), no. 3, 155–239.
- Alfred Hero and Bala Rajaratnam, Large-scale correlation screening, J. Amer. Statist. Assoc. 106 (2011), no. 496, 1540–1552. MR 2896855, DOI 10.1198/jasa.2011.tm11015
- Alfred Hero and Bala Rajaratnam, Hub discovery in partial correlation graphs, IEEE Trans. Inform. Theory 58 (2012), no. 9, 6064–6078. MR 2966078, DOI 10.1109/TIT.2012.2200825
- Ross Kindermann and J. Laurie Snell, Markov random fields and their applications, Contemporary Mathematics, vol. 1, American Mathematical Society, Providence, R.I., 1980. MR 620955
- Achim Klenke, Probability theory, Translation from the German edition, Universitext, Springer, London, 2014. A comprehensive course. MR 3112259, DOI 10.1007/978-1-4471-5361-0
- Steffen L. Lauritzen, Graphical models, Oxford Statistical Science Series, vol. 17, The Clarendon Press, Oxford University Press, New York, 1996. Oxford Science Publications. MR 1419991
- E. L. Lehmann, Elements of large-sample theory, Springer Texts in Statistics, Springer-Verlag, New York, 1999. MR 1663158, DOI 10.1007/b98855
- László Lovász and Balázs Szegedy, Limits of dense graph sequences, J. Combin. Theory Ser. B 96 (2006), no. 6, 933–957. MR 2274085, DOI 10.1016/j.jctb.2006.05.002
- Paul-A. Meyer, Probability and potentials, Blaisdell Publishing Co. [Ginn and Co.], Waltham, Mass.-Toronto, Ont.-London, 1966. MR 0205288
- Judea Pearl and Azaria Paz, GRAPHOIDS: A graph-based logic for reasoning about relevance relations, Tech. Report 8500038 (R-53), UCLA Computer Science Department, October 1985.
- Jonas Peters, On the intersection property of conditional independence and its application to causal discovery, preprint, arXiv:1403.0408 (2014).
- Frank Spitzer, Markov random fields and Gibbs ensembles, Amer. Math. Monthly 78 (1971), 142–154. MR 277036, DOI 10.2307/2317621
- Elias M. Stein and Rami Shakarchi, Complex analysis, Princeton Lectures in Analysis, vol. 2, Princeton University Press, Princeton, NJ, 2003. MR 1976398
- Martin P. Tingley and Peter Huybers, A Bayesian algorithm for reconstructing climate anomalies in space and time. Part II: Comparison with the regularized expectation maximization algorithm, J. Climate 23 (2010), no. 10, 2782–2800.
- A. M. Yaglom, Correlation theory of stationary and related random functions. Vol. I, Springer Series in Statistics, Springer-Verlag, New York, 1987. Basic results. MR 893393
- Fuzhen Zhang (ed.), The Schur complement and its applications, Numerical Methods and Algorithms, vol. 4, Springer-Verlag, New York, 2005. MR 2160825, DOI 10.1007/b105056
Additional Information
- David Montague
- Affiliation: Department of Statistics, University of California, Davis, California 95616
- MR Author ID: 902752
- Email: davmont@gmail.com
- Bala Rajaratnam
- Affiliation: Department of Statistics, University of California, Davis, California 95616
- MR Author ID: 861028
- Email: brajaratnam01@gmail.com
- Received by editor(s): February 4, 2015
- Received by editor(s) in revised form: August 14, 2016
- Published electronically: June 7, 2018
- Additional Notes: The authors were supported in part by the US NSF under grants DMS-CMG-1025465, AGS-1003823, DMS-1106642, and DMS-CAREER-1352656, and by the US Air Force Office of Scientific Research grant award FA9550-13-1-0043
- © Copyright 2018 American Mathematical Society
- Journal: Trans. Amer. Math. Soc. 370 (2018), 7557-7603
- MSC (2010): Primary 60G05, 60G15, 60G60, 60K35
- DOI: https://doi.org/10.1090/tran/7048
- MathSciNet review: 3852441