Pricing in General Insurance

512 Pricing in General Insurance

28.8 Appendix: R Code for Generating an Arbitrary Number

of Variables Correlated through a t Copula

Through the copula library and the related libraries, it is straightforward to produce cor-

related variables with R. In this appendix, we go straight to the most difcult case – that

of n statistical distributions (which could be severity, frequency, or aggregate loss distribu-

tions) correlated through a t copula. The distributions are given as empirical distributions

so that it is not necessary to assume any specic model (such as lognormal).

For increased readability, we have highlighted the code and the code comments ...