Využití kopul při modelování ve financích v kontextu současné regulatoriky

This thesis deals with the possibilities of using copulas in the area of risk management and measurement. Copulas represent a flexible tool for multivariate statistical modeling, allowing the construction of various forms of dependency between random variables. A possible area of use can be the calculation of the capital requirements required by the contemporary applied Solvency II legislation. The results of the work demonstrate the necessity to distinguish the character of the dependence between variables and thus copulas. This is due to the significant differences in the properties of the resulting joint distributions, which have an impact primarily on the quantiles and hence on the risk rates based on them. The disadvantage of the copula models is the need of the integration of simulation methods in the calculations of the properties of the join distribution. The development of internal models in individual insurance companies should include the application of couplas, at least in form of an identification of the relationship of the variables in order to assess whether there are significant differences between them for calculating the capital requirements compared to the standard formula.