ABSTRACT
Copula theory provides an easy way to deal with the (otherwise) complex multivariate modeling. e “… essential idea of the copula approach is that a joint distribution can be factored into the marginals and a dependence function called a copula. e term ‘copula’ means ‘link’: the copula couples the marginal distributions together in order to form a joint distribution. e dependence relationship is entirely determined by the copula, while scaling and shape (mean, standard deviation, skewness, and kurtosis) are entirely determined by the marginals …” (Rosenberg and Schuermann, 2004, p. 12). Copulas can therefore be used to obtain more realistic multivariate densities than the traditional joint Normal one, which is simply the product of a Normal copula and Normal marginals: for example, the Normal dependence relation can be preserved using a Normal copula, but marginals can be entirely general, e.g., Student’s t marginals.
