Hi all, I'm using the rmgarch package to model asset returns using the function cgarchfit function. The latest version of this package currently supports Gaussian copula and Student-t copula with time varying correlation rho_t. I have read the vignette but still not sure how the rho_t is modeled. There is, however, a brief mention of the seminal paper of Patton (2006) [1] on extension of static copulas to dynamic models, so is it safe to assume that the rho_t is modeled as per Patton?s approach? Or is it just a by-product of fitting a DCC-GARCH model a la Engle's? If anyone could clarify that for me it would be great, thank you! References [1] A.J. Patton. Modelling asymmetric exchange rate dependence. International Economic Review, 47(2):527?556, 2006. -Van
Copula-GARCH with rmgarch package
1 message · Le Hoang Van