Header

UZH-Logo

Maintenance Infos

Density and Risk Prediction with Non-Gaussian COMFORT Models


Paolella, Marc S; Polak, Paweł (2023). Density and Risk Prediction with Non-Gaussian COMFORT Models. Annals of Financial Economics, 18(01):2250033.

Abstract

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

Abstract

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

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Banking and Finance
Dewey Decimal Classification:330 Economics
Scopus Subject Areas:Social Sciences & Humanities > Business and International Management
Social Sciences & Humanities > Finance
Social Sciences & Humanities > Economics and Econometrics
Uncontrolled Keywords:GJR-GARCH, Multivariate generalized hyperbolic distribution, Non-ellipticity, Value-at-risk
Scope:Discipline-based scholarship (basic research)
Language:English
Date:March 2023
Deposited On:02 Oct 2023 13:24
Last Modified:30 Mar 2024 04:44
Publisher:World Scientific Publishing
ISSN:2010-4952
Additional Information:Bereits als Working Paper in SSRN No. 4280472 erschienen: https://dx.doi.org/10.2139/ssrn.4280472
OA Status:Closed
Publisher DOI:https://doi.org/10.1142/s2010495222500336
Related URLs:https://www.zora.uzh.ch/id/eprint/232027/
https://dx.doi.org/10.2139/ssrn.4280472
Other Identification Number:merlin-id:24081
Full text not available from this repository.