Practical Notes On Multivariate Modeling Based on Elliptical Copulas

Wang, Xiaojing; Yan, Jun

Journal de la société française de statistique, Tome 154 (2013) no. 1, pp. 102-115.

Les distributions multivariées construites à partir de copules elliptiques sont utilisées dans de nombreux domaines comme l’hydrologie et la finance. Nous nous intéressons à deux aspects pratiques concernant ces modèles. Dans un premier temps, nous attirons l’attention sur l’importance de la propriété de consistance définie par Kano (1994, Journal of Multivariate Analysis, 51 :139–147). Certaines distributions elliptiques ne satisfont pas cette propriété, ce qui limite les applications et l’implantation des copules correspondantes dans les logiciels. Dans un deuxième temps, nous donnons deux méthodes conditionnelles pour la génération d’échantillons aléatoires à partir de distributions elliptiques. La première approche présentée est fondée sur une représentation stochastique des distributions elliptiques alors que la seconde utilise une méthode d’acceptation/rejet. L’utilisation des deux méthodes est illustrée dans la cadre de la modélisation de données hydrologiques trivariées.

Multivariate distributions based on elliptical copulas have been widely used in many fields such as hydrology and finance. We focus on two practical issues of applications of such models. The first is a caveat rooted in a consistency property defined by Kano (1994, Journal of Multivariate Analysis, 51:139–147) for elliptical distributions. Some elliptical families do not have this property, which puts practical limitations on applications and software implementation of the corresponding elliptical copulas. The second issue is on conditional sampling from such distributions, which is important in Monte Carlo statistical inferences, especially when closed-form solutions are not available or feasible. Two sampling methods are presented: a direct sampling approach based on a stochastic representation of elliptical distributions, and an acceptance/rejection sampling method. The latter also provides an importance sampler as a byproduct, which may have higher efficiency for some applications. A trivariate model of the volume, duration, and peak intensity of annual extreme storms illustrates the sampling algorithms.

Keywords: conditional sampling, elliptical distribution, importance sampling, marginal consistency
Mot clés : distribution elliptique, génération conditionnelle d’échantillons aléatoires, propriété de consistance

[1] Abdous, Belkacem; Genest, Christian; Rémillard, Bruno Dependence Properties of Meta-Elliptical Distributions, Statistical Modeling and Analysis for Complex Data Problems (Duchesne, Pierre; Rémillard, Bruno, eds.), Springer US, 2005, pp. 1-15

[2] Bilodeau, Martin Comments on ‘Tail Conditional Expectations for Elliptical Distributions’, North American Actuarial Journal, Volume 7 (2003), pp. 118-122 | Zbl

[3] Basu, Sankar; Micchelli, Charles A; Olsen, Peder Power Exponential Densities for the Training and Classification of Acoustic Feature Vectors in Speech Recognition, Journal of Computational and Graphical Statistics, Volume 10 (2001) no. 1, pp. 158-184

[4] Cambanis, Stamatis; Huang, Steel; Simons, Gordon On the Theory of Elliptically Contoured Distributions, Journal of Multivariate Analysis, Volume 11 (1981), pp. 368-385 | Zbl

[5] Cherubini, Umberto; Luciano, Elisa; Vecchiato, Walter Copula Methods in Finance, John Wieley & Son Ltd, 2004, 293 pages | Zbl

[6] Devroye, L. Non-uniform Random Variate Generation, Springer-Verlag Inc, 1986, 843 pages | Zbl

[7] Fang, Hong-Bin; Fang, Kai-Tai; Kotz, Samuel The Meta-elliptical Distributions with Given Marginals, Journal of Multivariate Analysis, Volume 82 (2002) no. 1, pp. 1-16 | Zbl

[8] Frahm, Gabriel; Junker, Markus; Szimayer, Alexander Elliptical Copulas: Applicability and Limitations, Statistics & Probability Letters, Volume 63 (2003) no. 3, pp. 275-286 | Zbl

[9] Fang, K.-T.; Kotz, S.; Ng, K.-W. Symmetric Multivariate and Related Distributions, Chapman and Hall, 1990 | Zbl

[10] Frees, Edward W.; Valdez, Emiliano A. Understanding Relationships Using Copulas, North American Actuarial Journal, Volume 2 (1998) no. 1, pp. 1-25 | Zbl

[11] Genz, Alan; Bretz, Frank Computation of Multivariate Normal and $t$ Probabilities, Lecture Notes in Statistics, 195, Springer-Verlage, Heidelberg, 2009 | Zbl

[12] Genz, Alan; Bretz, Frank; Miwa, Tetsuhisa; Mi, Xuefei; Leisch, Friedrich; Scheipl, Fabian; Hothorn, Torsten mvtnorm: Multivariate Normal and $t$ Distributions (2011) http://CRAN.R-project.org/package=mvtnorm (R package version 0.9-9991)

[13] Gilks, W. R.; Best, N. G.; Tan, K. K. C. Adaptive Rejection Metropolis Sampling within Gibbs Sampling (Corr: 97V46 P541-542 with R. M. Neal), Applied Statistics, Volume 44 (1995), pp. 455-472 | Zbl

[14] Genest, C.; Favre, A.-C. Everything You Always Wanted to Know about Copula Modeling but Were Afraid to Ask, Journal of Hydrological Engineering, Volume 12 (2007), pp. 347-368

[15] Genest, C.; Favre, A. C.; Béliveau, J.; Jacques, C. Metaelliptical Copulas and Their Use in Frequency Analysis of Multivariate Hydrological Data, Water Resources Research, Volume 43 (2007) no. 9, pp. 1-12

[16] Genest, Christian; MacKay, Jock The Joy of Copulas: Bivariate Distributions with Uniform Marginals (Com: 87V41 P248), The American Statistician, Volume 40 (1986), pp. 280-283

[17] Gómez-Sánchez-Manzano, E.; Gómez-Villegas, M. A.; Marín, J. M. Multivariate Exponential Power Distributions as Mixtures of Normal Distributions with Bayesian Applications, Communications in Statistics – Theory and Methods, Volume 37 (2008) no. 6, pp. 972-985 | Zbl

[18] Hashorva, E. Tail Asymptotic Results for Elliptical Distributions, Insurance: Mathematics and Economics, Volume 43 (2008), pp. 158-164 | Zbl

[19] Hofert, M. Sampling Archimedean Copulas, Computational Statistics & Data Analysis, Volume 52 (2008) no. 12, pp. 5163-5174 | Zbl

[20] Hofert, M. Efficiently Sampling Nested Archimedean Copulas, Computational Statistics & Data Analysis, Volume 55 (2011) no. 1, pp. 57-70 | Zbl

[21] Joe, H. Multivariate Models and Dependence Concepts, Chapman and Hall, London, 1997 | Zbl

[22] Kano, Yutaka Consistency Property of Elliptical Probability Density Functions, Journal of Multivariate Analysis, Volume 51 (1994), pp. 139-147 | Zbl

[23] Kao, S. C.; Govindaraju, R. S. A Bivariate Frequency Analysis of Extreme Rainfall with Implications for Design, Journal of Geophysical Research. D. Atmospheres, Volume 112 (2007) | DOI

[24] Kojadinovic, I.; Yan, J. Comparison of Three Semiparametric Methods for Estimating Dependence Parameters in Copula Models, Insurance: Mathematics and Economics, Volume 47 (2010) no. 1, pp. 52-63 | Zbl

[25] Kojadinovic, I.; Yan, J. Modeling Multivariate Distributions with Continuous Margins Using the copula R Package, Journal of Statistical Software, Volume 34 (2010) no. 9, pp. 1-20

[26] Lindsey, J. K. Multivariate Elliptically Contoured Distributions for Repeated Measurements, Biometrics, Volume 55 (1999) no. 4, p. 1277-80 | Zbl

[27] Landsman, Zinoviy M.; Valdez, Emiliano A. Tail Conditional Expectations for Elliptical Distributions, North American Actuarial Journal, Volume 7 (2003), pp. 55-71 | Zbl

[28] Marsaglia, G. Choosing a Point from the Surface of a Sphere, Annals of Mathematical Statistics, Volume 43 (1972), pp. 645-646 | Zbl

[29] McNeil, A. J.; Frey, R.; Embrechts, P. Quantitative risk management, Princeton Series in Finance, Princeton University Press, New Jersey, 2005 | Zbl

[30] Manner, Hans; Segers, Johan Tails of Correlation Mixtures of Elliptical Copulas, Insurance: Mathematics and Economics, Volume 48 (2011) no. 1, pp. 153-160 | Zbl

[31] Nelsen, R.B. An Introduction to Copulas, Springer, New-York, 2006 (Second edition) | Zbl

[32] Nolan, J. Multivariate Elliptically Contoured Stable Distributions: Theory and Estimation (2006) (Technical Report)

[33] Petris, Giovanni; Tardella, Luca; Gilks, Wally HI: Simulation from Distributions Supported by Nested Hyperplanes (2006) (R package version 0.3)

[34] Robert, Christian P.; Casella, George Monte Carlo Statistical Methods, Springer-Verlag, 2004, 645 pages | Zbl

[35] Salvadori, G.; De Michele, C.; Kottegoda, N. T.; Rosso, R. Extremes in Nature: An Approach Using Copulas, Water Science and Technology Library Series, Springer, 2007

[36] Sklar, A. Fonctions de répartition à $n$ dimensions et leurs marges, Publications de l’Institut de Statistique de l’Université de Paris, Volume 8 (1959), pp. 229-231 | Zbl

[37] Venables, W. N.; Ripley, B. D. Modern Applied Statistics with S, Springer, New York, 2002 http://www.stats.ox.ac.uk/pub/MASS4 (ISBN 0-387-95457-0) | Zbl

[38] Wang, X.; Gebremichael, M.; Yan, J. Weighted Likelihood Copula Modeling of Extreme Rainfall Events in Connecticut, Journal of Hydrology, Volume 390 (2010) no. 1–2, pp. 108-115