Abstract
The purpose of this paper is to present a comprehensive Monte Carlo simulation study on the performance of minimum-distance (MD) and maximum-likelihood (ML) estimators for bivariate parametric copulas. In particular, I consider Cramér-von-Mises-, Kolmogorov-Smirnov- and L 1-variants of the CvM-statistic based on the empirical copula process, Kendall’s dependence function and Rosenblatt’s probability integral transform. The results presented in this paper show that regardless of the parametric form of the copula, the sample size or the location of the parameter, maximum-likelihood yields smaller estimation biases at less computational effort than any of the MD-estimators. The MD-estimators based on copula goodness-of-fit metrics, on the other hand, suffer from large biases especially when used for estimating the parameters of archimedean copulas. Moreover, the results show that the bias and efficiency of the minimum-distance estimators are strongly influenced by the location of the parameter. Conversely, the results for the maximum-likelihood estimator are relatively stable over the parameter interval of the respective parametric copula.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Ané T, Kharoubi C (2003) Dependence structure and risk measure. J Bus 76(3): 411–438
Berg D (2009) Copula goodness-of-fit testing: an overview and power comparison. Eur J Fin 15(7): 675–701
Biau G, Wegkamp M (2005) Minimum distance estimation of copula densities. Stat Probab Lett 73: 105–114
Breymann W, Dias A, Embrechts P (2003) Dependence structures for multivariate high-frequency data in finance. Quant Fin 3: 1–14
Chen X, Fan Y (2006) Estimation of copula-based semiparametric time series models. J Econ 130: 307–335
Chen S, Poon S (2007) Modelling international stock market contagion using copula and risk appetite. Working paper
Deheuvels P (1978) Caractérisation complète des Lois Extrèmes Multivariées et de la Convergence des Types Extrèmes. Pub l’Institut Stat l’Université Paris 23: 1–36
Deheuvels P (1981) A nonparametric test for independence. Pub l’Institut Stat l’Université Paris
Demarta S, McNeil A (2005) The t-copula and related copulas. Int Stat Rev 73(1): 111–129
Durrleman V, Nikeghbali A, Roncalli T (2000) Which copula is the right one? Groupe Rech Oper Crédit Lyonnais: Working paper
Embrechts P, McNeil A, Straumann D (2002) Correlation and dependency in risk management: properties and pitfalls. Risk Manag: Value Risk Beyond, Dempster M, Univ Press Cambridge, Cambridge, pp 176–223
Fermanian J (2005) Goodness-of-fit tests for copulas. J Multivar Anal 95: 119–152
Fermanian J, Radulovic D, Wegkamp M (2004) Weak convergence of empirical copula processes. Bernoulli 10: 847–860
Genest C, Mackay RJ (1986) Copules arquimédiennes et familles des lois bidimensionelles dont les marges sont données. Can J Stat 14: 145–159
Genest C, Rivest LP (1993) Statistical inference proceduresfor bivariate Archimedean copulas. J Am Stat Assoc 88: 1034–1043
Genest C, Ghoudi K, Rivest L (1995) A semiparametric estimation procedure of dependence parameters in multivariate families of distributions. Biometrika 82: 543–552
Genest C, Quessy JF, Rémillard B (2006) Goodness-of-fit procedures for copula models based on the integral probability transformation. Scand J Stat 33: 337–366
Genest C, Rémillard B, Beaudoin D (2009) Goodness-of-fit tests for copulas: A review and a power study. Insur Math Econ 44: 199–213
Ghoudi C, Rémillard B (2004) Empirical processes based on pseudo-observations. II. The multivariate case. Asymptot Methods Stoch Fields Inst Commun Am Math Soc Provid 44: 381–406
Hoeffding W (1940) Scale invariant correlation theory. Schr Math Inst Univ Berlin 5: 181–233
Huard D, Évin G, Favre AC (2006) Bayesian copula selection. Comp Stat Data Anal 51(2): 809–822
Hutchinson T, Lai C (1990) Continuous multivariate distributions. emphasising applications. Adelaide
Joe H (1997) Multivariate models and dependence concepts. London
Junker M, May A (2005) Measurement of aggregate risk with copulas. Econ J 8: 428–454
Kim G, Silvapulle M, Silvapulle P (2007) Comparison of semiparametric and parametric methods for estimating copulas. Comp Stat Data Anal 51: 2836–2850
Kole E, Koedijk K, Verbeek M (2007) Selecting copulas for risk management. J Bank Fin 31(8): 2405–2423
Li D (2000) On default correlation: a copula approach. J Fixed Income 9: 43–54
Malevergne Y, Sornette D (2003) Testing the Gaussian copula hypothesis for financial assets dependence. Quant Fin 3: 231–250
McNeil A, Frey R, Embrechts P (2005) Quantitative risk management. Princeton
Mendes B, De Melo B, Nelsen R (2007) Robust fits for copula models. Comm Stat Sim Comp 36(5): 997–1017
Nelsen R (2006) An introduction to copulas, 2nd edn. New York
Nikoloulopoulos A, Karlis D (2008) Copla model evaluation based on parametric bootstrap. Comp Stat Data Anal 52: 3342–3353
Resnick SI (1987) Extreme values, regular variation, and point processes. Berlin
Rodriguez J (2007) Measuring financial contagion: a copula approach. J Empir Fin 41: 401–423
Rosenblatt M (1952) Remarks on a multivariate transformation. Ann Math Stat 23: 470–472
Savu C, Trede M (2008) Goodness-of-fit tests for parametric families of archimedean copulas. Quant Fin 8: 109–116
Schmid F, Trede M (1996) An L 1-variant of the Cramer-von-Mises test. Stat Prob Lett 26: 91–96
Sibuya M (1960) Bivariate extreme statistics. Ann Inst Stat Math 11: 195–210
Sklar A (1959) Fonctions de repartition à n dimensions et leurs marges. Publ Inst Univ Paris 8: 229–231
Tsukahara H (2005) Semiparametric estimation in copula models. Can J Stat 33(3): 357–375
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Weiß, G. Copula parameter estimation by maximum-likelihood and minimum-distance estimators: a simulation study. Comput Stat 26, 31–54 (2011). https://doi.org/10.1007/s00180-010-0203-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-010-0203-7