PUB - Publikationen an der Universität Bielefeld

In the last decade, simplified vine copula models have been an active area of research. They build ahigh dimensional probability density from the product of marginals densities and bivariate copula densities. Besides parametric models, several approaches to nonparametric estimation of vine copulas have been proposed. In this article, we extend these approaches and compare them in an extensive simulation study and areal data application. We identify several factors driving the relative performance of the estimators. The mostimportant one is the strength of dependence. No method was found to be uniformly better than all others.Overall, the kernel estimators performed best, but do worse than penalized B-spline estimators when thereis weak dependence and no tail dependence.