Hierarchical Archimedean Copulae

Abstract

This paper aims at explanation of the R-package HAC, which provides user friendly methods for dealing with high-dimensional hierarchical Archimedean copulae (HAC). A computationally eficient estimation procedure allows to recover the structure and the parameters of HACs from data. In addition, arbitrary HACs can be constructed to sample random vectors and to compute the values of the corresponding cumulative distribution as well as density functions. Accurate graphics of the important characteristics of the package's object hac can be produced by the generic plot function.