Copula-based approaches, presented in section 3, seem adequate and
preferable to the widely employed assumption of multivariate Gaussian distributions4
of risk factor changes, if the risk factor changes are not normally
distributed.5 Section 3.1 gives an introduction to copula-based approaches
in the context of top-down risk aggregation. Various parametric distribution
functions that are used to model marginal distributions in the context of risk
aggregation are shortly mentioned in section 3.2. Section 3.3 presents some
selected bivariate and two multi-dimensional copulas in detail and compares
their properties. Specific equations for copula functions and copula densities
are also provided in this section. Different approaches to copula parameter
estimation are presented in section 3.4, and goodness-of-fit tests are presented
in section 3.5. Finally, section 3.6 provides algorithms for the simulation of
the presented copulas.