Estimators based on Kendall’s tau in multivariate copula models. (English) Zbl 1274.62367

Genest, Christian; Nešlehová, Johanna; Ben Ghorbal, Noomen

Aust. N. Z. J. Stat. 53, No. 2, 157-177 (2011).

Summary: The estimation of a real-valued dependence parameter in a multivariate copula model is considered. Rank-based procedures are often used in this context to guard against possible misspecification of the marginal distributions. A standard approach consists of maximizing the pseudo-likelihood. Here, we investigate alternative estimators based on the inversion of two multivariate extensions of Kendall’s tau developed by Kendall and Babington Smith, and by Joe. The former, which amounts to the average value of tau over all pairs of variables, is often referred to as the coefficient of agreement. Existing results concerning the finite- and large-sample properties of this coefficient are summarized, and new, parallel findings are provided for the multivariate version of tau due to Joe, along with illustrations. The performance of the estimators resulting from the inversion of these two versions of Kendall’s tau is compared in the context of copula models through simulations.

MSC:

62H12	Estimation in multivariate analysis
62H20	Measures of association (correlation, canonical correlation, etc.)

Keywords:

Archimedean copula; coefficient of agreement; finite-sample moments; Kendall’s tau; meta-elliptical copula

Software:

QRM

References:

[1]	Abdous, Statistical Modeling and Analysis for Complex Data Problems pp 1– ((2005)) · doi:10.1007/0-387-24555-3_1
[2]	Barbe, On Kendall’s process, J. Multivariate Anal. 58 pp 197– ((1996)) · Zbl 0862.60020 · doi:10.1006/jmva.1996.0048
[3]	Cherubini, Copula Methods in Finance ((2004)) · doi:10.1002/9781118673331
[4]	Daniels, The significance of rank correlations where parental correlation exists, Biometrika 34 pp 197– ((1947)) · Zbl 0030.20801 · doi:10.1093/biomet/34.3-4.197
[5]	Ehrenberg, On sampling from a population of rankers, Biometrika 39 pp 82– ((1952)) · Zbl 0046.35901 · doi:10.1093/biomet/39.1-2.82
[6]	El Maache, Mathematical Statistics and Applications: Festschrift for Constance van Eeden pp 113– ((2003))
[7]	Fang, The meta-elliptical distributions with given marginals, J. Multivariate Anal. 94 pp 222– ((2002)) · doi:10.1016/j.jmva.2004.10.001
[8]	Fermanian, Capital Formation, Governance and Banking pp 59– ((2005))
[9]	Frees, Understanding relationships using copulas, North Amer. Act. J. 2 pp 1– ((1998)) · Zbl 1081.62564 · doi:10.1080/10920277.1998.10595667
[10]	Genest, A semiparametric estimation procedure of dependence parameters in multivariate families of distributions, Biometrika 82 pp 543– ((1995)) · Zbl 0831.62030 · doi:10.1093/biomet/82.3.543
[11]	Hays, A note on average tau as a measure of concordance, J. Amer. Statist. Assoc. 55 pp 331– ((1960)) · Zbl 0212.22403 · doi:10.2307/2281746
[12]	Hoeffding, On the distribution of the rank correlation coefficient {\(\tau\)} when the variates are not independent, Biometrika 34 pp 183– ((1947))
[13]	Hoeffding, A class of statistics with asymptotically normal distribution, Ann. Math. Statist. 19 pp 293– (1948) · Zbl 0032.04101 · doi:10.1214/aoms/1177730196
[14]	Hult, Multivariate extremes, aggregation and dependence in elliptical distributions, Adv. Appl. Probab. 34 pp 587– (2002) · Zbl 1023.60021 · doi:10.1239/aap/1033662167
[15]	Joe, Multivariate concordance, J. Multivariate Anal. 35 pp 12– (1990) · Zbl 0741.62061 · doi:10.1016/0047-259X(90)90013-8
[16]	Jouini, Copula models for aggregating expert opinions, Operations Res. 44 pp 444– (1996) · Zbl 0864.90067 · doi:10.1287/opre.44.3.444
[17]	Kendall, Rank Correlation Methods (1990) · Zbl 0732.62057
[18]	Kendall, On the method of paired comparisons, Biometrika 31 pp 324– (1940) · Zbl 0023.14803 · doi:10.1093/biomet/31.3-4.324
[19]	Kim, Comparison of semiparametric and parametric methods for estimating copulas, Comput. Statist. Data Anal. 51 pp 2836– (2007) · Zbl 1161.62364 · doi:10.1016/j.csda.2006.10.009
[20]	Klüppelberg, Copula structure analysis, J. Roy. Statist. Soc. Ser. B 71 pp 737– (2009) · Zbl 1250.62031 · doi:10.1111/j.1467-9868.2009.00707.x
[21]	Kojadinovic, Comparison of three semiparametric methods for estimating dependence parameters in copula models, Insurance Math. Econom. 47 pp 52– (2010) · Zbl 1231.62044 · doi:10.1016/j.insmatheco.2010.03.008
[22]	Kotz, Continuous Multivariate Distributions: Models and Applications (2000) · Zbl 0946.62001 · doi:10.1002/0471722065
[23]	Lindeberg, Den VI skandinaviske Matematikerkongres i København, 31 August-4 September 1925 pp 437– (1927)
[24]	Lindeberg, Some remarks on the mean error of the percentage of correlation, Nordic Statist. J. 1 pp 137– (1929) · JFM 55.0921.01
[25]	McNeil, Quantitative Risk Management: Concepts, Techniques, and Tools (2005) · Zbl 1089.91037
[26]	McNeil, Multivariate Archimedean copulas, d-monotone functions and 1-norm symmetric distributions, Ann. Statist. 37 pp 3059– (2009) · Zbl 1173.62044 · doi:10.1214/07-AOS556
[27]	McNeil, From Archimedean to Liouville copulas, J. Multivariate Anal. 101 pp 1772– (2010) · Zbl 1190.62102 · doi:10.1016/j.jmva.2010.03.015
[28]	Nelsen, Distributions with Fixed Marginals and Related Topics (Seattle, WA, 1993) pp 223– (1996) · doi:10.1214/lnms/1215452621
[29]	Oakes, A model for association in bivariate survival data, J. Roy. Statist. Soc. Ser. B 44 pp 414– (1982) · Zbl 0503.62035
[30]	Salvadori, Extremes in Nature: An Approach Using Copulas (2007)
[31]	Schucany, Correlation structure in Farlie-Gumbel-Morgenstern distributions, Biometrika 65 pp 650– (1978) · Zbl 0397.62033 · doi:10.1093/biomet/65.3.650
[32]	Shih, Inferences on the association parameter in copula models for bivariate survival data, Biometrics 51 pp 1384– (1995) · Zbl 0869.62083 · doi:10.2307/2533269
[33]	Sklar, Fonctions de répartition àn dimensions et leurs marges, Publ. Inst. Statist. Univ. Paris 8 pp 229– (1959)
[34]	Sundrum, Moments of the rank correlation coefficient {\(\tau\)} in the general case, Biometrika 40 pp 409– (1953) · Zbl 0051.36904
[35]	Tiit, Distributions with Fixed Marginals and Related Topics (Seattle, WA, 1993) pp 337– (1996) · doi:10.1214/lnms/1215452630