Smooth Nonparametric Bernstein Vine Copulas

35 Pages Posted: 30 Sep 2012 Last revised: 8 Oct 2012

See all articles by Gregor N. F. Weiss

Gregor N. F. Weiss

University of Leipzig - Faculty of Economics and Management Science

Marcus Scheffer

University of Dortmund - Economics and Social Sciences

Date Written: October 7, 2012

Abstract

We propose the use of nonparametric Bernstein copulas as bivariate pair-copulas in high-dimensional vine models. The resulting smooth and nonparametric vine copulas completely obviate the error-prone need for choosing the pair-copulas from parametric copula families. By means of a simulation study and an empirical analysis of financial market data, we show that our proposed smooth nonparametric vine copula model is superior to competing parametric vine models calibrated via Akaike’s Information Criterion.

Keywords: finance, dependence structures, vine copulas, Bernstein copulas

JEL Classification: C52, C53, C58

Suggested Citation

Weiss, Gregor N. F. and Scheffer, Marcus, Smooth Nonparametric Bernstein Vine Copulas (October 7, 2012). Available at SSRN: https://ssrn.com/abstract=2154458 or http://dx.doi.org/10.2139/ssrn.2154458

Gregor N. F. Weiss (Contact Author)

University of Leipzig - Faculty of Economics and Management Science ( email )

Grimmaische Str. 12
Leipzig, 04109
Germany
+49 341 97 33821 (Phone)
+49 341 97 33829 (Fax)

HOME PAGE: http://www.wifa.uni-leipzig.de/nfdl

Marcus Scheffer

University of Dortmund - Economics and Social Sciences ( email )

United States