Last modified: 2018-08-29
Abstract
English version:
The PC algorithm is the most popular algorithm used to infer the structure of a Bayesian network directly from data. For Gaussian distributions, it infers the network structure using conditional independence tests based on Pearson correlation coefficients. Here, we propose two modified versions of PC, the R-vine PC and D-vine PC algorithms, suitable for elliptical copula data. The correlation matrix is inferred by means of the estimated structure and parameters of a regular vine. Simulation results are provided, showing the very good performance of the proposed algorithms with respect to their main competitors.
Â
References
1. Aas, K., Czado, C., Frigessi, A., Bakken, H.: Pair-copula constructions of multiple dependence. Insurance: Mathematics and economics 44(2), 182–198 (2009)
2. Bauer, A., Czado, C.: Pair-copula bayesian networks. J. Comput. Graph. Stat. 25(4), 1248– 1271 (2016)
3. Bedford, T., Cooke, R.M.: Probability density decomposition for conditionally dependent random variables modeled by vines. Ann. Math. Artif. Intell. 32(1), 245–268 (2001)
4. Bedford, T., Cooke, R.M.: Vines: A new graphical model for dependent random variables. Ann. Stat. 30(4), 1031–1068 (2002)
5. Brechmann, E.C., Czado, C., Aas, K.: Truncated regular vines in high dimensions with application to financial data. Can. J. Stat. 40(1), 68–85 (2012)
6. Cowell, R.G., Dawid, P., Lauritzen, S.L., Spiegelhalter, D.J.: Probabilistic Networks and Expert Systems. Springer-Verlag, New York. (1999)
7. Dissmann, J., Brechmann, E.C., Czado, C., Kurowicka, D.: Selecting and estimating regular vine copulae and application to financial returns. Comput. Stat. Data. Anal. 59, 52–69 (2013)
8. Elidan, G.: Copula bayesian networks. In: Advances in neural information processing systems, pp. 559–567 (2010)
9. Genest, C., Favre, A.: Everything you always wanted to know about copula modeling but were afraid to ask. J. Hydrol. Eng. 12(4), 347–368 (2007)
10. Hanea, A.M., Kurowicka, D., Cooke, R.M.: Hybrid method for quantifying and analyzing bayesian belief nets. Qual. Reliab. Eng. Int. 22(6), 709–729. (2006)
11. Hobæk Haff, I., Aas, K., Frigessi, A., Lacal, V.: Structure learning in bayesian networks using regular vines. Comput. Stat. Data Anal. 101(C), 186–208 (2016)
12. Joe, H.: Families of m-variate distributions with given margins and m (m-1)/2 bivariate dependence parameters. Lecture Notes-Monograph Series pp. 120–141 (1996)
13. Kalisch, M., Buhlmann, P.: Estimating high-dimensional directed acyclic graphs with the pc-algorithm. J. Mach. Learn. Res. 8, 613–636. (2007)
14. Kurowicka, D., Cooke, R.: Uncertainty Analysis with High Dimensional Dependence Modelling. John Wiley & Sons, Ltd. (2006)
15. Naftali, H., Drton, M.: Pc algorithm for nonparanormal graphical models. J. Mach. Learn. Res. 14, 3365–3383. (2013)
16. Neapolitan, R.E.: Learning Bayesian Networks. Prentice-Hall, Inc., Upper Saddle River, NJ, USA. (2003)
17. Sklar, A.: Fonctions de répartition à n dimensions et leurs marges. Publications de l’Institut de Statistique de L’Université de Paris ´ 8, 229–231. (1959)
18. Spirtes, P., Glymour, C., Scheines, R.: Causation, Prediction, and Search, 2nd edn. MIT press, Cambridge, Massachusetts. (2000)
19. Tsamardinos, I., Brown, L.E., Aliferis, C.F.: The max-min hill-climbing bayesian network structure learning algorithm. Mach. Learn. 65(1), 31–78 (2006)