| Full text | |
| Author(s): |
Total Authors: 3
|
| Affiliation: | [1] Univ Carlos III, Dept Stat, Calle Madrid 126, Getafe 28903, Madrid - Spain
[2] Univ Estadual Campinas, Rua Sergio Buarque de Holanda 651, Campinas, SP - Brazil
Total Affiliations: 2
|
| Document type: | Journal article |
| Source: | INTERNATIONAL JOURNAL OF FORECASTING; v. 34, n. 1, p. 45-63, JAN-MAR 2018. |
| Web of Science Citations: | 5 |
| Abstract | |
Multivariate GARCH (MGARCH) models need to be restricted so that their estimation is feasible in large systems and so that the covariance stationarity and positive definiteness of conditional covariance matrices are guaranteed. This paper analyzes the limitations of some of the popular restricted parametric MGARCH models that are often used to represent the dynamics observed in real systems of financial returns. These limitations are illustrated using simulated data generated by general VECH models of different dimensions in which volatilities and correlations are interrelated. We show that the restrictions imposed by the BEKK model are very unrealistic, generating potentially misleading forecasts of conditional correlations. On the other hand, models based on the DCC specification provide appropriate forecasts. Alternative estimators of the parameters are important in order to simplify the computations, and do not have implications for the estimates of conditional correlations. The implications of the restrictions imposed by the different specifications of MGARCH models considered are illustrated by forecasting the volatilities and correlations of a five dimensional system of exchange rate returns. (C) 2017 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved. (AU) | |
| FAPESP's process: | 13/00506-1 - Time series, wavelets and functional data analysis |
| Grantee: | Pedro Alberto Morettin |
| Support Opportunities: | Research Projects - Thematic Grants |