Riccomagno, Eva and Smith, J. Q., 1953- (2005) The causal manipulation and Bayesian estimation of chain event graphs. Working Paper. University of Warwick. Centre for Research in Statistical Methodology, Coventry.

Preview

PDF WRAP_Riccomagno_05-16w.pdf - Published Version - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Download (514Kb)

Abstract

Discrete Bayesian Networks (BNs) have been very successful as a framework both for inference and for expressing certain causal hypotheses. In this paper we present a class of graphical models called the chain event graph (CEG) models, that generalises the class of discrete BN models. This class is suited for representing conditional independence and sample space structures of asymmetric models. It retains many useful properties of discrete BNs, in particular admitting conjugate estimation. It provides a flexible and expressive framework for representing and analysing the implications of causal hypotheses, expressed in terms of the effects of a manipulation of the generating underlying system.We prove that, as for a BN, identifiability analyses of causal effects can be performed through examining the topology of the CEG graph, leading to theorems analogous to the Backdoor theorem for the BN.

Item Type:	Working or Discussion Paper (Working Paper)
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Science > Statistics
Library of Congress Subject Headings (LCSH):	Graphical modeling (Statistics), Bayesian statistical decision theory
Series Name:	Working papers
Publisher:	University of Warwick. Centre for Research in Statistical Methodology
Place of Publication:	Coventry
Date:	2005
Volume:	Vol.2005
Number:	No.16
Number of Pages:	36
Status:	Not Peer Reviewed
Access rights to Published version:	Open Access
References:	[1] P. Anderson and J.Q. Smith (2005). A graphical framework for represent- ing the semantics of asymmetric models. Technical report 05-12, CRiSM, Department of Statistics, The University of Warwick. [2] T. Bedford and R. Cooke (2001). Probabilistic risk analysis: foundations and methods. Cambridge University Press, Cambridge. [3] C. Boutilier, N. Freidman, M. Goldszmidt and D. Koller (1996). Context- specific independence in Bayesian networks. In Proceedings of UAI - 96 115-123. [4] R.E. Bryant (1986). Graphical algorithms for Boolean function manipu- lation. IEEE Transactions of Computers C-35 677-691. [5] R.J. Castelo (2002). The Discrete Acyclic Digraph Markov Model in Data Mining. Utrecht University, Ph.D. thesis. [6] R. Castelo and M.D. Perlman (2004). Learning essential graph Markov models from data. In A. Gamze, S. Moral and A. Salmeron (eds.) Ad- vances in Bayesian networks, Springer, Berlin, 255–269. [7] G.A. Churchill (1995). Accurate restoration of DNA sequences. In C. Gatsaris et al. (eds.) Case Studies in Bayesian Statistics vol. II, Springer- Verlag, New-York, 90-148. [8] D. Cooper and C. Yoo (1999). Causal discovery from a mixture of ex- perimental and observational data. In K.B. Laskey and H. Prade (eds.) Proceedings of the Seventeen Conference on Uncertainty in Artificial In- telligence, Morgan Kaufmann, San Francisco. [9] R.G. Cowell, A.P. Dawid, S.L. Lauritzen and D.J. Spiegelhalter (1999). Probabilistic Networks and Expert Systems, Springer-Verlag, New York. [10] A.P. Dawid (2000). Causality without counterfactuals J. Amer. Statist. Ass. 95:407-448. [11] D.G.T. Denison, C.C. Holmes, B.K. Mallick and A.F.M. Smith (2002). Bayesian methods for nonlinear classification and regression, John Wiley & Sons Ltd., Chichester. [12] N. Freidman and M. Goldszmidt (1999). Learning Bayesian networks with local structure. In M.I. Jordan (ed.) Learning in Graphical Models MIT Press, 421-459. [13] S. French (ed.) (1989). Readings in Decision analysis. Chapman and Hall/CRC, London. [14] D. Glymour and G.F. Cooper (1999). Computation, Causation, and Dis- covery. MIT Press, Cambridge, MA. [15] D. Hausman (1998). Causal Asymmetries, Cambridge University Press, Cambridge. [16] P.W. Holland (1986). Statistics and causal inference (With discussion and a reply by the author). Journal of the American Statistical Association, 81(396):945–970. [17] M.I. Jordan (ed.) (1999). Learning in Graphical Models, MIT Press, Cam- bridge, MA. [18] M. Jaeger (2004). Probabilistic decision graphs - combining verification and AI techniques for probabilistic inference. Int.J. of Uncertainty, Fuzzi- ness and Knowledge-based Systems, 12:19-42. [19] S.L. Lauritzen (1996). Graphical models. Oxford Science Press, Oxford, 1st edition. [20] R. Lyons (1990). Random walks and percolation on trees. Annals of Prob- ability, 18:931-958. [21] D. Madigan and A. E. Raftery (1994). Model selection and accounting for model uncertainty in graphical models using Occam’s window. Journal of the American Statistical Association 89(428): 1535-1546. [22] A.M. Madrigal and J.Q. Smith (2004). Causal Identification in Design Networks. In Sucar et al. (eds.) Advances in Artificial Intelligence 2, Springer-Verlag, 517-526. [23] A.M. Madrigal (2004). Evaluations of Policy Interventions under Exper- imental Conditions using Bayesian Influence Diagrams. The University of Warwick, Ph.D. Thesis. [24] D. McAllester, M. Collins and F. Pereira (2004). Case-factor diagrams for structured probability models. In Proceedings of UAI - 2004 382-391. [25] J. Pearl (1988). Probabilistic Reasoning in Intelligent Systems. Morgan Kauffman, San Mateo. [26] J. Pearl (1995). Causal diagrams for empirical research. Biometrika, 82:669-710. [27] J. Pearl (2000). Causality. models, reasoning and inference. Cambridge University Press, Cambridge. [28] J. Pearl (2003). Statistics and Causal Inference: A Review (with discus- sion). Test, 12(2):281-345. [29] D. Poole and N.L. Zhang (2003). Exploiting contextual independence in probabilistic inference. Journal of Artificial Intelligence research, 18:263- 313. [30] E. Riccomagno and J.Q. Smith (2003). Non-Graphical Causality: a gen- eralisation of the concept of a total cause. Research report series No. 394, Dept of Statistics, The University of Warwick. [31] E. Riccomagno and J.Q. Smith (2004). Identifying a cause in models which are not simple Bayesian networks. In Proceedings of the 10th Con- ference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, 1315 -1322. [32] J.M. Robins (1986). A new approach to causal inference in mortality studies with a sustained exposure period —application to control of the healthy worker survivor effect. Mathematical Modeling, 7:1393-1512. [33] J.M. Robins (1997). Causal inference from complex longitudinal data. In M. Berkane (ed.) Latent variable modeling and applications to causality (Los Angeles, CA, 1994). Springer-Verlag, New York, 69–117. [34] J.M. Robins, R. Scheines, P. Spirtes, and L. Wasserman (2003). Uniform Consistency in Causal Inference. Biometrika 90(3):491-515. [35] D. Rubin (1973). Estimating causal effects of treatments in randomised and non - randomised studies. J. Educational Psychology 66:688-701. [36] G. Shafer (1996). The Art of Causal Conjecture. MIT Press, Cambridge, MA. [37] R. Settimi and J.Q. Smith (1998). On the geometry of Bayesian graph- ical models with hidden variables. In G. Cooper and S. Moral (eds.) Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann Publishers, S. Francisco, 472–479. [38] R. Settimi and J.Q. Smith (2000). Geometry, moments and conditional independence trees with hidden variables. The Annals of Statistics, 28(4):1179-1205. [39] J.Q. Smith, A.E. Faria, S. French, D. Ranyard, J. Bohunova, T. Dura- nova, M. Stubna, L. Dutton, C. Rojas and A. Sohier (1997). Probabilistic data assimilation within RODOS. Radiation Protection Dosimetry 73(1- 4):57-59. [40] J.Q. Smith and E. E. Anderson (2006). Conditional independence and Chain Event Graphs. Artificial Intelligence, to appear. [41] J.Q. Smith and J. Croft (2003). Bayesian networks for discrete multivari- ate data: An algebraic approach to inference. J. of Multivariate Analysis, 84(2):387-402. [42] D. Spiegelhalter, A.P. Dawid, S.L. Lauritzen and R.G. Cowell (1993). Bayesian analysis of expert systems. Statistical Science, 8:219-282. [43] P. Spirtes, C. Glymour and R. Scheines (1993). Causation, Prediction, and Search. Springer-Verlag, New York.
URI:	http://wrap.warwick.ac.uk/id/eprint/35590

Request changes to a record

Actions (login required)

View Item