We provide the first and second derivatives of (log-) densities and conditional
distribution functions of various bivariate copulas.
These derivatives are required in order to calculate e.g. the observed Fisher
information in multivariate models based on bivariate copulas.
In particular, we obtain all derivatives for the bivariate t-copula, which have not
been available until now.
All derivatives are implemented in the R package VineCopula, and we
demonstrate the accuracy of our implementation by comparing
the Fisher information matrices calculated using our functions with known analytical
results where available.
«
We provide the first and second derivatives of (log-) densities and conditional
distribution functions of various bivariate copulas.
These derivatives are required in order to calculate e.g. the observed Fisher
information in multivariate models based on bivariate copulas.
In particular, we obtain all derivatives for the bivariate t-copula, which have not
been available until now.
All derivatives are implemented in the R package VineCopula, and we
demonstrate the accuracy of our implem...
»