Checking dependence structure between variables is important when modeling
multivariate data. In this paper, we investigate copulas, functions that provide a
flexible methodology for modeling multivariate dependence. In particular, copulas
are useful in describing the dependence structure of extreme values in the tails.
Gauss, Student t- and Cauchy copulas are considered, among others. We also
investigate a method for the choice of copulas. Based on the chosen copula,
we demonstrate a simple graphical method with data from a study on multiple
myeloma, particularly for analysis of the tail dependence.