Bivariate Wavelet-Based Method for Estimating Copula Function
Nargess Hosseinioun
Statistics, University, Mashhad, Khorasan Razavi, Islamic Republic of Iran

In the recent years, the copula models became a popular tool for modeling dependencies between random variables, especially in such fields as biostatistics, actuarial science, and finance.The copulas model has been extensively studied in a parametrical frame-work for the distribution function c. Large classes of copulas, such as the elliptic family, which contains the Gaussian copula and the Student copula, and the Archimedian family, which contains the Gumbel copula, the Clayton copula and the Frank copulas, have been identified.

The aim of this work is to establish an upper bound on L p$'$-losses (2 ‚ȧ p$'$ < 1) of the linear wavelet-based estimator for copula function when the copula function assumed to be bouned and the marginals are unknown and must be estimated from ranks.

References:

[1] Antoniadis, A. (2007), Wavelet methods in statistics: some recent developments and their applications. Statist. Surv., 1, 16-55.

[2] Antoniadis, A., Gregoire, Mc. Keague (1994). Wavelet methods for curve estimation, American Statistical Association, Vol. 89, No. 428.

[3] Autin, F., Le Pennec, E., and Tribouley, K. (2010). Thresholding method to estimate the copula density. Journal of Multivariate Analysis, 101, pp. 200-222.

[4] Bergh J., Loftstorm J. (1976): Interpolation spaces: An introduction. Springer.

[5] Biau, G. and Wegkamp, M. H. (2005). A note on minimum distance estimation of copula densities. Statist. Probab. Lett., 73, 105-114.

[6]Breaks, R. and keilegom, I. Van (2009). Flexible modeling based on copulas in nonparametric regression. J. Multiv. Anal., 6, 1270-1281.

[7] Chen, X. and Fan, Y. (2005). Pseudo-likelihood ratio tests for semiparametric mul- tivariate copula model selection. Canad. J. Statist., 33, 389-414.

[8] Donoho, D.L., Johnstone, I.M., Kerkyacharian, G., Picard, D. (1995). Wavelet shrinkage: asymptopia (with discussion),J. Roy. Statist. Soc. Ser. B 57(2), 301-369.8

[9] Doosti, H., Chaubey, Y.P. (2007). Wavelet linear density estimation for negatively dependent random variables, Current development in theory and applications of wavelets, 1, Issue 1, 57-64.

[10] Dearden, L., Fitzsimons, E. and Goodman, A. (2006). Estimating Lifetime Earnings Distributions Using Copulas, IFS Working paper.

[11] Fermanian, J.-D. and Scaillet, O. (2005). Some statistical pitfalls in copula modelling for financial applications. In Klein, E., editor, Capital Formation, Governance and Banking, pages 59-74. Nova Science Publishing, New York.

[12] Genest, C., Masiello, E. and Tribouley, K. (2009). Estimating copula densities through wavelets. Mathematics and Economics, 44, 170-181

[13] Gijbels, I. and Mielniczuk, J. (1990). Estimating the density of a copula function. Comm. Statist. Theory Methods, 19, 445-464.

[14] Morettin, Pedro. A., Toloi, M. C., Chiann, C. and Miranda, J.(2010). Wavelet Smoothed Empirical Copula Estimators. Revista Brasileira de Finanas, 8, 263-281.

[15] Peetre J. (1976) New thoughts on Besov spaces. Duke University, Mathematics Series 1.

Keywords: Copula; Rank statistics; Wavelet

Biography: I have finished my PhD in Statistics in Ferdowsi University of Mashad-Iran last year. I started working as an asistant professor in PNU university last year.

I have a particular interest in Nonparametric estimation, wavelets with applications in Biostatistics and Finance.