Higher-dimensional Copula Models and Their Application: Bayesian Inference for D-vine Pair-copula Constructions Based on Different Bivariate Families

Modelling multivariate dependence structures has proven to be an important aspect in the field of finance. The reality of financial markets shows clear evidence that asset returns exhibit non-normal dependence. Since copula functions have been applied to the solution of these non-normal distributed problems, they find wide-ranging application in the fields of risk management, derivative pricing, hedging and optimal portfolio decisions. Pair-copula construction allows modelling of the dependence structure between different higher-dimensional time series. The author shows a flexible way to estimate and calibrate such higher-dimensional copula models and provides the application of U. S. industrial returns. In his simulation study he shows the sensitivity with respect to the copula parameter and different families using newly developed R-packages.