Constructing hierarchical archimedean copulas with Lévy subordinators (Q968494): Difference between revisions

Regular Archimedean copulas exhibit exchangeability of dependence structure which is often a not realistic property. Hierarchical Archimedean copulas are usually presented as a way around to this drawback. This article deals with parametric hierarchical Archimedean copulas given by independent exponential random variables divided by group specific Lévy subordinators which are evaluated at a common random time. Hence, the resulting vector has specific groups of components presenting distinct dependence structures. Moreover, the approach offers an efficient sampling algorithm which is of easy construction.