Publication: Analyzing the effect of introducing a kurtosis parameter in Gaussian Bayesian networks
Loading...
Full text at PDC
Publication Date
2009-05
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier Sci. Ltd.
Citations
Exportar
Abstract
Gaussian Bayesian networks are graphical models that represent the dependence structure of a multivariate normal random variable with a directed acyclic graph (DAG). In Gaussian Bayesian networks the output is usually the conditional distribution of some unknown variables of interest given a set of evidential nodes whose values are known. The problem of uncertainty about the assumption of normality is very common in applications. Thus a sensitivity analysis of the non-normality effect in our conclusions could be necessary. The aspect of non-normality to be considered is the tail behavior. In this line, the multivariate exponential power distribution is a family depending on a kurtosis parameter that goes from a leptokurtic to a platykurtic distribution with the normal as a mesokurtic distribution. Therefore a more general model can be considered using the multivariate exponential power distribution to describe the joint distribution of a Bayesian network, with a kurtosis parameter reflecting deviations from the normal distribution. The sensitivity of the conclusions to this perturbation is analyzed using the Kullback-Leibler divergence measure that provides an interesting formula to evaluate the effect.
Description
UCM subjects
Unesco subjects
Keywords
Citation
Pearl J. Probabilistic reasoning in intelligent systems. Morgan Publishers, Inc.; 1988.
Jensen FV. An introduction to Bayesian networks. NY: Springer; 1996.
Lauritzen SL, Spiegelhalter DJ. Local computation with probabilities in graphical structures and their applications to expert systems. J R Stat Soc B 1988;50(2):154–227.
Li Z, D’Ambrosio B. Efficient inference in Bayes networks as a combinatorial optimization problem. Int J Approx Reason 1994;11:55–81.
Batchelor C, Cain J. Application of belief networks to water management studies. Agric Water Manage 1999;40(1):51–7.
Cowell R. FINEX: a probabilistic expert system for forensic identification. Forensic Sci Int 2003;134:196–206.
Friedman N, et al. Using Bayesian networks to analyze expression data. In: Proceedings of the fourth annual international conference on computational molecular biology; 2000.
Langseth L, Portinale L. Bayesian networks in reliability. Reliab Eng Syst Safety 2007;92(1):92–108.
Gómez E, Gómez-Villegas MA, Marín JM. A multivariate generalization of the power exponential family of distributions. Commun Stat 1998;B27:589–600.
Kullback S, Leibler R. On information and sufficiency. Ann Stat 1951;22: 79–86.
Box GEP, Tiao GC. Bayesian inference in statistical analysis. New York, NY: Wiley; 1992.
Castillo E, Gutie´ rrez JM, Hadi AS. Sensitivity analysis in discrete Bayesian networks. IEEE Trans Syst Man Cybern 1997;26(7):412–23.
Laskey KB. Sensitivity analysis for probability assesments in Bayesian networks. IEEE Trans Syst Man Cybern 1995;25:412–23.
Castillo E, Kjærulff U. Sensitivity analysis in Gaussian Bayesian networks using a symbolic-numerical technique. Reliab Eng Syst Safety 2003;79(2): 139–48.
Gómez-Villegas MA, Main P, Susi R. Sensitivity analysis in Gaussian Bayesian networks using a divergence measure. Commun Stat 2007;B36:523–39.
Abramowitz M, Stegun I. Handbook of mathematical functions with formulas, graphs, and mathematical tables. In: Abramowitz M, Stegun IA, editors. Reprint of the 1972 edition. New York, NY: Dover Publications, Inc; 1992.