Testing for Structural Breaks in Correlations: Does it Improve Value-at-Risk Forecasting?

39 Pages Posted: 17 May 2013 Last revised: 16 Mar 2015

See all articles by Tobias Berens

Tobias Berens

University TU Dortmund; zeb/rolfes.schierenbeck.associates

Gregor N. F. Weiss

University of Leipzig - Faculty of Economics and Management Science

Dominik Wied

University of Cologne

Date Written: March 11, 2015

Abstract

In this paper, we compare the Constant Conditional Correlation (CCC) model to its dynamic counterpart, the Dynamic Conditional Correlation (DCC) model with respect to its accuracy for forecasting the Value-at-Risk of financial portfolios. Additionally, we modify these benchmark models by combining them with a pairwise test for constant correlations, a test for a constant correlation matrix, and a test for a constant covariance matrix. In an empirical horse race of these models based on five- and ten-dimensional portfolios, our study shows that the plain CCC- and DCC-GARCH models are outperformed in several settings by the approaches modified by tests for structural breaks in asset correlations and covariances.

Keywords: CCC-GARCH, DCC-GARCH, Estimation window, Structural breaks, VaR-forecast

JEL Classification: C32, C41, C53, G17, G32

Suggested Citation

Berens, Tobias and Berens, Tobias and Weiss, Gregor N. F. and Wied, Dominik, Testing for Structural Breaks in Correlations: Does it Improve Value-at-Risk Forecasting? (March 11, 2015). Journal of Empirical Finance, Forthcoming, Available at SSRN: https://ssrn.com/abstract=2265488 or http://dx.doi.org/10.2139/ssrn.2265488

Tobias Berens

University TU Dortmund ( email )

Emil-Figge-Straße 50
Dortmund, 44227
Germany

zeb/rolfes.schierenbeck.associates ( email )

Hammer Straße 165
Münster, 48149
Germany

Gregor N. F. Weiss (Contact Author)

University of Leipzig - Faculty of Economics and Management Science ( email )

Grimmaische Str. 12
Leipzig, 04109
Germany
+49 341 97 33821 (Phone)
+49 341 97 33829 (Fax)

HOME PAGE: http://www.wifa.uni-leipzig.de/nfdl

Dominik Wied

University of Cologne ( email )

Albertus-Magnus-Platz
Cologne, 50923
Germany

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
405
Abstract Views
2,597
Rank
132,693
PlumX Metrics