Density and Risk Prediction with Non-Gaussian COMFORT Models

The CCC-GARCH model, and its dynamic correlation extensions, form the most important model class for multivariate asset returns. For density portfolio risk forecasting, a drawback of these models is underlying assumption Gaussianity. This paper considers so-called COMFORT class, which but endowed with generalized hyperbolic innovations. novelty that parameter estimation conducted by joint maximum likelihood, all parameters, using an EM algorithm, so feasible hundreds assets. demonstrates (i) new blatantly superior to Gaussian counterpart in terms forecasting ability, (ii) also outperforms ad-hoc three-step procedures common literature augment CCC DCC fat-tailed distribution. An extensive empirical study confirms model’s superiority Value-at-Risk forecasting.