|
Copulas have become an important element of best practice for enterprise risk management, displacing in many contexts other more conservative approaches to modelling stochastic dependence. An `ideal' copula should conform to a wide range of problems at hand, being either symmetric or asymmetric, and exhibiting flexible extent of tail dependence. The so-called full-range tail dependence copulas are exactly such a model, and thus they have recently been proved very useful for modeling various dependence patterns in the joint distributional tails.
An inconvenient aspect of the full-range tail dependence copulas is the computational speed, which is acceptable but may not be fast enough for complex applications of modeling high-dimensional dependence structures. In this project, we propose an interpolation approach to overcome the issue. The practical usefulness of our approach will be illustrated with financial data.
This material is based upon work supported by the National Science Foundation under Grant No. 1246818.
|
