Abstract
Bayesian networks are powerful tools for handling problems which are specified through a multivariate probability distribution. A broad background of theory and methods have been developed for the case in which all the variables are discrete. However, situations in which continuous and discrete variables coexist in the same problem are common in practice. In such cases, usually the continuous variables are discretized and therefore all the existing methods for discrete variables can be applied, but the price to pay is that the obtained model is just an approximation. In this chapter we study two frameworks where continuous and discrete variables can be handled simultaneously without using discretization. These models are based on the CG and MTE distributions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
E. Castillo, J. Gutiérrez, A. Hadi, Expert systems and probabilistic network models, Springer-Verlag, New York, 1997.
B. Cobb, P. Shenoy, R. Rumí, Approximating probability density functions with mixtures of truncated exponentials, in: Proceedings of Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU-2004), In press, Perugia, Italy, 2004.
B. Cobb, P. Shenoy, Inference in hybrid Bayesian networks with mixtures of truncated exponentials, International Journal of Approximate Reasoning, 2006, In press.
R. Cowell, A. Dawid, S. Lauritzen, D. Spiegelhalter, Probabilistic Networks and Expert Systems, Series: Statistics for Engineering and Information Science, Springer, 1999.
A. Dempster, N. Laird, D. Rubin, Maximun likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society B 39 (1977) 1–38.
M. El-Taha, W. Evans, A new estimation procedure for the right-truncated exponential distribution, in: Proceedings of the 23th Pittsburgh Conference on Modelling and Simulation, 1992, pp. 427–434.
Elvira Consortium, Elvira: An environment for creating and using probabilistic graphical models, in: J. Gámez, A. Salmerón (Eds.), Proceedings of the First European Workshop on Probabilistic Graphical Models, 2002, pp. 222–230.
D. Geiger, D. Heckerman, Learning Gaussian networks, Proceedings of the 10th Conference on Uncertainty in Artificial Intelligence, Morgan & Kaufmann, 1994, 235–243.
W. Gilks, S. Richardson, D. Spiegelhalter, Markov chain Monte Carlo in practice, Chapman and Hall, London, UK, 1996.
L. Hegde, R. Dahiya, Estimation of the parameters of a truncated Gamma distribution, Communications in Statistics. Theory and Methods 18 (1989) 561–577.
F.V. Jensen, S.L. Lauritzen, K.G. Olesen, Bayesian updating in causal probabilistic networks by local computation, Computational Statistics Quarterly 4 (1990) 269–282.
D. Koller, U. Lerner, D. Anguelov, A general algorithm for approximate inference and its application to hybrid Bayes nets, Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence, Morgan & Kaufmann, 1999, 324–333.
D. Kozlov, D. Koller, Nonuniform dynamic discretization in hybrid networks, in: D. Geiger, P. Shenoy (Eds.), Proceedings of the 13th Conference on Uncertainty in Artificial Intelligence, Morgan & Kaufmann, 1997, pp. 302–313.
S. Kullback, S., R.A. Leibler, On information and sufficiency, Annals of Mathematical Statistics 22 (1951) 79–86.
S.L. Lauritzen, D.J. Spiegelhalter, Local computations with probabilities on graphical structures and their application to expert systems, Journal of the Royal Statistical Society, Series B 50 (1988) 157–224.
S.L. Lauritzen, N. Wermuth, Graphical models for associations between variables, some of which are qualitative and some quantitative, The Annals of Statistics 17 (1989) 31–57.
S. Lauritzen, Propagation of probabilities, means and variances in mixed graphical association models, Journal of the American Statistical Association 87 (1992) 1098–1108.
S. Lauritzen, F. Jensen, Stable local computation with conditional Gaussian distributions, Statistics and Computing 11 (2001) 191–203.
H. Leimer, Triangulated graphs with marked vertices, Annals of Discrete Mathematics 41, 1989, pp. 311–324.
U. Lerner, E. Segal, D. Koller, Exact inference in networks with discrete children of continuous parents, Proceedings of the 17th Conference on Uncertainty in Artificial Intelligence, Morgan & Kaufmann, 2001, pp. 319–328.
A.L. Madsen, F.V. Jensen, Lazy propagation: a junction tree inference algorithm based on lazy evaluation, Artificial Intelligence 113 (1999) 203–245.
S. Moral, R. Rumí, A. Salmerón, Mixtures of truncated exponentials in hybrid Bayesian networks. ECSQARU’01, Lecture Notes in Artificial Intelligence 2143 (2001) 135–143.
S. Moral, R. Rumí, A. Salmerón, Estimating mixtures of truncated exponentials from data, in: J. Gámez, A. Salmerón (Eds.), Proceedings of the First European Workshop on Probabilistic Graphical Models, 2002, pp. 156–167.
S. Moral, R. Rumí, A. Salmerón, Approximating conditional MTE distributions by means of mixed trees. ECSQARU’03, Lecture Notes in Artificial Intelligence 2711 (2003) 173–183.
K. Murphy, Fitting a Conditional Linear Gaussian Distribution, Technical Report, 1998.
K. Murphy, A variational approximation for Bayesian networks with discrete and continuous latent variables, Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence, Morgan & Kaufmann, 1999, 467–475.
G. Nath, Unbiased estimates of reliability for the truncated Gamma distribution, Scandinavian Actuarial Journal 3 (1975) 181–186.
K.G. Olesen, Causal probabilistic networks with both discrete and continuous variables, IEEE Transactions on Pattern Analysis and Machine Intelligence 15 (1993) 275–279.
J. Quinlan, Induction of decision trees, Machine Learning 1 (1986) 81–106.
R. Redner, H. Walker, Mixture densities, maximum likelihood and the EM algorithm, SIAM Review 26 (1984) 195–239.
V. Romero, R. Rumí, A. Salmerón, Learning hybrid Bayesian networks with mixtures of truncated exponentials, International Journal of Approximate Reasoning (2005) In press.
R. Rumí, A. Salmerón, Penniless propagation with mixtures of truncated exponentials. ECSQARU’05. Lecture Notes in Computer Science 3571 (2005) 39–50.
R. Rumí, A. Salmerón, S. Moral, Estimating mixtures of truncated exponentials in hybrid Bayesian network, Test (2005) In press.
A. Salmerón, A. Cano, S. Moral, Importance sampling in Bayesian networks using probability trees, Computational Statistics and Data Analysis 34 (2000) 387–413.
Y. Sathe, S. Varde, Minimum variance unbiased estimates of reliability for the truncated exponential distribution, Technometrics 11 (1969) 609–612.
P.P. Shenoy, G. Shafer, Axioms for probability and belief function propagation, in: R.D. Shachter, T.S. Levitt, J.F. Lemmer and L.N. Kanal (eds.) Uncertainty in Artificial Intelligence, vol. 4. North Holland, Amsterdam, 1990, pp. 169–198.
J. Simonoff, Smoothing methods in Statistics, Springer, 1996.
W. Smith, A note on truncation and sufficient statistics, Annals of Mathematical Statistics 28 (1957) 247–252.
J. Tukey, Sufficiency, truncation and selection, Annals of Mathematical Statistics 20 (1949) 309–311.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Cobb, B.R., Rumí, R., Salmerón, A. (2007). Bayesian Network Models with Discrete and Continuous Variables. In: Lucas, P., Gámez, J.A., Salmerón, A. (eds) Advances in Probabilistic Graphical Models. Studies in Fuzziness and Soft Computing, vol 213. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-68996-6_4
Download citation
DOI: https://doi.org/10.1007/978-3-540-68996-6_4
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-68994-2
Online ISBN: 978-3-540-68996-6
eBook Packages: EngineeringEngineering (R0)