zbMATH — the first resource for mathematics

A new class of continuous Bayesian networks. (English) Zbl 07119658
Summary: In this paper, we introduce a new family of continuous Bayesian networks whose conditional distributions belong to the Tweedie class. This family of distributions is very important in statistical modeling as it includes a variety of density shapes and is widely used in many real-world applications in various fields. It also includes many well-known probability distributions such as Gaussian, Gamma and Inverse-Gaussian. In this context, the constructed Tweedie Bayesian network (TBN), which is based on Tweedie regression models, constitutes a wide and flexible class of Bayesian networks. A full procedure is provided for the learning of the TBN structure and parameters. We also introduce a sensitivity measure relying on the Kullback-Leibler divergence in order to perform a sensitivity analysis and validate the model output. The proposed graphical model is illustrated through both an application to a loan subscribers data set and a simulation study.
68T37 Reasoning under uncertainty in the context of artificial intelligence
Full Text: DOI
[1] Pearl, J., Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, (1988), Morgan Kaufmann Publishers Inc.
[2] Jarraya, A.; Leray, P.; Masmoudi, A., Discrete exponential Bayesian networks: definition, learning and application for density estimation, Neurocomputing, 137, 142-149, (2014)
[3] Neapolitan, R. E., Learning Bayesian Networks, (2003), Pearson Prentice Hall: Pearson Prentice Hall Upper Saddle River, NJ
[4] Shachter, R. D.; Kenley, C. R., Gaussian influence diagrams, Manag. Sci., 35, 5, 527-550, (1989)
[5] Grzegorczyk, M., An introduction to Gaussian Bayesian networks, (Systems Biology in Drug Discovery and Development, (2010), Springer), 121-147
[6] Jöreskog, K. G.; Wold, H. O., Systems Under Indirect Observation: Causality, Structure, Prediction, (1982), North Holland · Zbl 0488.00020
[7] Yang, E.; Ravikumar, P.; Allen, G. I.; Liu, Z., Graphical models via univariate exponential family distributions, J. Mach. Learn. Res., 16, 1, 3813-3847, (2015) · Zbl 1351.62111
[8] Jørgensen, B., The Theory of Dispersion Models, (1997), CRC Press · Zbl 0928.62052
[9] Jørgensen, B., Exponential dispersion models (with discussion), J. R. Stat. Soc. B, 49, 127-162, (1987) · Zbl 0662.62078
[10] Tweedie, M., An index which distinguishes between some important exponential families, (Statistics: Applications and New Directions: Proc. Indian Statistical Institute Golden Jubilee International Conference, vol. 579, (1984)), 579-604
[11] Smyth, G. K.; Jørgensen, B., Fitting Tweedie’s compound Poisson model to insurance claims data: dispersion modelling, ASTIN Bull., J. IAA, 32, 1, 143-157, (2002) · Zbl 1094.91514
[12] Kendal, W. S.; Lagerwaard, F. J.; Agboola, O., Characterization of the frequency distribution for human hematogenous metastases: evidence for clustering and a power variance function, Clin. Exp. Metastasis, 18, 3, 219-229, (2000)
[13] Hassine, A.; Masmoudi, A.; Ghribi, A., Tweedie regression model: a proposed statistical approach for modelling indoor signal path loss, Int. J. Numer. Model., 30, 6, Article e2243 pp., (2017)
[14] Vinogradov, V., On the power-variance family of probability distributions, Commun. Stat., Theory Methods, 33, 5, 1007-1029, (2004) · Zbl 1114.60308
[15] Jørgensen, B., Generalized Linear Models, Encyclopedia of Environmetrics, 1152-1159, (2013)
[16] Nelder, J. A.; Baker, R. J., Generalized Linear Models, (1972), Wiley Online Library
[17] Bonat, W. H.; Kokonendji, C. C., Flexible Tweedie regression models for continuous data, J. Stat. Comput. Simul., 87, 11, 2138-2152, (2017)
[18] Fornberg, B.; Sloan, D. M., A review of pseudospectral methods for solving partial differential equations, Acta Numer., 3, 203-267, (1994) · Zbl 0808.65117
[19] Gourieroux, C.; Monfort, A.; Trognon, A., Pseudo maximum likelihood methods: theory, Econometrica, 52, 3, 681-700, (1984) · Zbl 0575.62031
[20] Jørgensen, B.; Knudsen, S. J., Parameter orthogonality and bias adjustment for estimating functions, Scand. J. Stat., 31, 1, 93-114, (2004) · Zbl 1051.62022
[21] Holst, R.; Jørgensen, B., Generalized linear longitudinal mixed models with linear covariance structure and multiplicative random effects, Chil. J. Stat. (ChJS), 6, 1, 5-36, (2015)
[22] McCullagh, P., Quasi-likelihood functions, Ann. Stat., 11, 1, 59-67, (1983) · Zbl 0507.62025
[23] Mooney, C. Z., Monte Carlo Simulation, vol. 116, (1997), Sage Publications
[24] Aistleitner, C.; Hofer, M.; Tichy, R., A central limit theorem for Latin hypercube sampling with dependence and application to exotic basket option pricing, Int. J. Theor. Appl. Finance, 15, 07, Article 1250046 pp., (2012) · Zbl 1255.91424
[25] Fahrmeir, L.; Kaufmann, H., Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models, Ann. Stat., 13, 1, 342-368, (1985) · Zbl 0594.62058
[26] Benjamini, Y.; Hochberg, Y., Controlling the false discovery rate: a practical and powerful approach to multiple testing, J. R. Stat. Soc., Ser. B, Methodol., 57, 1, 289-300, (1995) · Zbl 0809.62014
[27] Abid, L.; Zaghdene, S.; Masmoudi, A.; Ghorbel, S. Z., Bayesian network modeling: a case study of credit scoring analysis of consumer loans default payment, Asian Econ. Financ. Rev., 7, 9, 846-857, (2017)
[28] Dunn, P., Tweedie Exponential Family Models, (2013), R package version 2.1.7
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.