Archimedean Copulas in High Dimensions: Estimators and Numerical Challenges Motivated by Financial Applications

Hofert, Marius; Mächler, Martin; McNeil, Alexander J.

Journal de la société française de statistique, Tome 154 (2013) no. 1, pp. 25-63.

Les pratiques du moment en gestion quantitative du risque nous amènent à étudier des modèles de dépendance en grandes dimensions construits à partir de copules Archimédiennes. La performance d’un grand nombre d’estimateurs, dont plusieurs sont originaux, est étudiée et des solutions sont proposées pour les problèmes numériques associés. Des estimateurs fondés sur le méthode des moments et le beta de Blomqvist ainsi que le tau de Kendall sont comparés à des estimateurs du minimum de distance, du maximum de vraisemblance, du maximum de vraisemblance simulé et du maximum de vraisemblance fondé sur la diagonale de la copule. Des simulations sont réalisées dans le cas où les marges sont connues et dans le cas où elles sont estimées non paramétriquement, en faibles et en grandes dimensions, pour plusieurs familles de copules Archimédiennes. Toutes les méthodes étudiées sont implantées dans le package R copula distribué sous une licence libre. Les solutions numériques présentées dans ce travail restent valident dans le cas de généralisations asymétriques des copules Archimédiennes et d’importantes quantités associées comme la fonction de distribution de Kendall.

The study of Archimedean dependence models in high dimensions is motivated by current practice in quantitative risk management. The performance of known and new parametric estimators for the parameters of Archimedean copulas is investigated and related numerical difficulties are addressed. In particular, method-of-moments-like estimators based on pairwise Kendall’s tau, a multivariate extension of Blomqvist’s beta, minimum distance estimators, the maximum-likelihood estimator, a simulated maximum-likelihood estimator, and a maximum-likelihood estimator based on the copula diagonal are studied. Their performance is compared in a large-scale simulation study both under known and unknown margins (pseudo-observations), in small and high dimensions, under small and large dependencies, and various different Archimedean families. High dimensions up to one hundred are considered and computational problems arising from such large dimensions are addressed in detail. All methods are implemented in the open source R package copula and can thus be easily accessed and studied. The numerical solutions developed in this work extend to various asymmetric generalizations of Archimedean copulas and important quantities such as distributions of radial parts or the Kendall distribution function.

Keywords: Archimedean copulas, parameter estimation, Kendall’s tau, Blomqvist’s beta, minimum distance estimators, (diagonal/simulated) maximum-likelihood estimation, quantitative risk management
Mot clés : Copules Archimédiennes, estimation paramétrique, tau de Kendall, beta de Blomqvist, estimateurs du minimum de distance, estimateurs du maximum de vraisemblance, gestion quantitative du risque

[1] Berg, D.; Aas, K. Models for construction of multivariate dependence – A comparison study, The European Journal of Finance, Volume 15 (2009) no. 7, pp. 639-659

[2] Berg, D. Copula goodness-of-fit testing: an overview and power comparison, The European Journal of Finance, Volume 15 (2009) no. 7-8, pp. 675-701 | DOI

[3] Barbe, P.; Genest, Christian; Ghoudi, K.; Rémillard, B. On Kendall’s Process, Journal of Multivariate Analysis, Volume 58 (1996), pp. 197-229 | MR | Zbl

[4] Brahimi, B.; Necir, A. A semiparametric estimation of copula models based on the method of moments (2011) (Accessed 2011-08-06) | arXiv | MR

[5] Chavez-Demoulin, Valérie; Embrechts, Paul; Hofert, Marius An extreme value approach for modeling operational risk losses depending on covariates (2013)

[6] Charpentier, A.; Fermanian, J.-D.; Scaillet, O. The Estimation of Copulas: Theory and Practice, Copulas: From Theory to Applications in Finance (Rank, J., ed.), Risk Books, 2007, pp. 35-62

[7] Donnelly, C.; Embrechts, P. The devil is in the tails: actuarial mathematics and the subprime mortgage crisis, ASTIN Bulletin, Volume 40 (2010) no. 1, pp. 1-33 | MR | Zbl

[8] Duchateau, L.; Janssen, P. The Frailty Model, Springer, 2008 | MR | Zbl

[9] Dimitrova, D. S.; Kaishev, V. K.; Penev, S. I. GeD spline estimation of multivariate Archimedean copulas, Computational Statistics & Data Analysis, Volume 52 (2008), pp. 3570-3582 | MR | Zbl

[10] D’Agostino, R. B.; Stephens, M. A. Goodness-of-fit techniques, Dekker, 1986 | MR | Zbl

[11] Embrechts, Paul; Hofert, Marius Comments on: Inference in multivariate Archimedean copula models, TEST, Volume 20 (2011) no. 2, pp. 263-270 | DOI | MR | Zbl

[12] Feller, W. An Introduction to Probability Theory and Its Applications, 2, Wiley, 1971 | MR | Zbl

[13] Flajolet, P.; Sedgewick, R. Mellin transforms and asymptotics: Finite differences and Rice’s integrals, Theoretical Computer Science, Volume 144 (1995), pp. 101-124 | MR | Zbl

[14] Genest, Christian; Ghoudi, K.; Rivest, L.-P. A semiparametric estimation procedure of dependence parameters in multivariate families of distributions, Biometrika, Volume 82 (1995) no. 3, pp. 543-552 | MR | Zbl

[15] Genest, Christian; Nešlehová, Johanna; Ziegel, Johanna Inference in multivariate Archimedean copula models, TEST (2011) (in press) | MR | Zbl

[16] Genest, Christian; Rivest, L.-P. Statistical Inference Procedures for Bivariate Archimedean Copulas, Journal of the American Statistical Association, Volume 88 (1993) no. 423, pp. 1034-1043 | MR | Zbl

[17] Genest, Christian; Rémillard, B.; Beaudoin, D. Goodness-of-fit tests for copulas: A review and a power study, Insurance: Mathematics and Economics, Volume 44 (2009), pp. 199-213 | MR | Zbl

[18] Griewank, Andreas; Walther, Andrea Introduction to Automatic Differentiation, Proceedings of GAMM 2002, Volume 2 (2003) no. 1, pp. 45-49 | Zbl

[19] Hering, C. Estimation Techniques and Goodness-of-fit Tests for Certain Copula Classes in Large Dimensions, Ulm University (2011) (Ph. D. Thesis)

[20] Hering, Christian; Hofert, Marius Goodness-of-fit tests for Archimedean copulas in large dimensions (2013)

[21] Hofert, Marius; Mächler, Martin Nested Archimedean Copulas Meet R: The nacopula Package, Journal of Statistical Software, Volume 39 (2011) no. 9, pp. 1-20 http://www.jstatsoft.org/v39/i09/ (Accessed 2011-08-06)

[22] Hofert, Marius; Mächler, Martin; McNeil, Alexander J. Likelihood inference for Archimedean copulas in high dimensions under known margins, Journal of Multivariate Analysis, Volume 110 (2012), pp. 133-150 | DOI | MR | Zbl

[23] Hofert, Marius Sampling Nested Archimedean Copulas with Applications to CDO Pricing, Südwestdeutscher Verlag für Hochschulschriften AG & Co. KG, 2010 (PhD thesis)

[24] Hofert, Marius A stochastic representation and sampling algorithm for nested Archimedean copulas, Journal of Statistical Computation and Simulation, Volume 82 (2012) no. 9, pp. 1239-1255 | DOI | MR | Zbl

[25] Hougaard, Philip Analysis of Multivariate Survival Data, Springer, 2000 | MR | Zbl

[26] Hofert, Marius; Pham, David Densities of nested Archimedean copulas, Journal of Multivariate Analysis (2013) | DOI | MR | Zbl

[27] Hofert, Marius; Scherer, Matthias CDO pricing with nested Archimedean copulas, Quantitative Finance, Volume 11 (2011) no. 5, pp. 775-787 | DOI | MR | Zbl

[28] Hofert, Marius; Vrins, Frédéric Sibuya copulas, Journal of Multivariate Analysis, Volume 114 (2013), pp. 318-337 | DOI | MR | Zbl

[29] Joe, H. Multivariate Models and Dependence Concepts, Chapman & Hall/CRC, 1997 | MR | Zbl

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

[31] Lambert, P. Archimedean copula estimation using Bayesian splines smoothing techniques, Computational Statistics & Data Analysis, Volume 51 (2007), pp. 6307-6320 | MR | Zbl

[32] McNeil, A. J.; Frey, R.; Embrechts, P. Quantitative Risk Management: Concepts, Techniques, Tools, Princeton University Press, 2005 | MR | Zbl

[33] McNeil, A. J.; Nešlehová, Johanna Multivariate Archimedean copulas, $d$-monotone functions and $l_1$-norm symmetric distributions, The Annals of Statistics, Volume 37 (2009) no. 5b, pp. 3059-3097 | MR | Zbl

[34] Mächler, Martin Numerically Stable Frank Copula Functions via Multiprecision: R Package ‘Rmpfr’, 2011 http://cran.r-project.org/web/packages/copula/vignettes/Frank-Rmpfr.pdf (Accessed 2012-07-01)

[35] Mächler, Martin Accurately Computing $log(1-exp(-\left|a\right|\left)\right)$ , 2012 http://cran.r-project.org/web/packages/Rmpfr/vignettes/log1mexp-note.pdf (Accessed 2012-06-28)

[36] Nelsen, R. B. An Introduction to Copulas, Springer, 2006 | MR | Zbl

[37] Qu, X.; Zhou, J.; Shen, X. Archimedean copula estimation and model selection via $l_1$-norm symmetric distribution, Insurance: Mathematics and Economics, Volume 46 (2010), pp. 406-414 | MR | Zbl

[38] Scarsini, M. On measures of concordance, Stochastica, Volume 8 (1984) no. 3, pp. 201-218 | MR | Zbl

[39] Sibuya, M. Bivariate extreme statistics, I, Annals of the Institute of Statistical Mathematics, Volume 11 (1959) no. 2, pp. 195-210 | MR | Zbl

[40] Schönbucher, P. J.; Schubert, D. Copula-Dependent Default Risk in Intensity Models (2001) http://papers.ssrn.com/sol3/papers.cfm?abstract_id=301968 (Accessed 2009-12-30)

[41] Schmid, F.; Schmidt, R. Nonparametric inference on multivariate versions of Blomqvist’s beta and related measures of tail dependence, Metrika, Volume 66 (2007), pp. 323-354 | MR | Zbl

[42] Savu, C.; Trede, M. Hierarchies of Archimedean copulas, Quantitative Finance, Volume 10 (2010) no. 3, pp. 295-304 | MR | Zbl

[43] Stephenson, A. G. High-dimensional parametric modelling of multivariate extreme events, Australian & New Zealand Journal of Statistics, Volume 51 (2009) no. 1, pp. 77-88 | MR | Zbl

[44] Tsukahara, H. Semiparametric estimation in copula models, The Canadian Journal of Statistics, Volume 33 (2005) no. 3, pp. 357-375 | MR | Zbl

[45] Weiß, G. N. F. Copula parameter estimation: numerical considerations and implications for risk management, The Journal of Risk, Volume 13 (2010) no. 1, pp. 17-53