Prognose von Zinsvolatilitäten mit Regime-Switching-Modellen: Eine empirische Analyse des Euro-DM-Geldmarktes
JOURNAL ARTICLE
Cite JOURNAL ARTICLE
Style
Format
Prognose von Zinsvolatilitäten mit Regime-Switching-Modellen: Eine empirische Analyse des Euro-DM-Geldmarktes
Credit and Capital Markets – Kredit und Kapital, Vol. 31 (1998), Iss. 3 : pp. 370–399
Additional Information
Article Details
Author Details
Ralf Ahrens, Gießen
References
-
Bera, Anil K./Higgins, Matthew L. (1993): ARCH Models: Properties, Estimation and Testing, Journal of Economic Surveys, Vol. 7, S. 305-366.
Google Scholar -
Bollerslev, Tim (1986): Generalized Autoregressive Conditional Heteroskedasticity, Journal of Econometrics, Vol. 31, S. 307-327.
Google Scholar -
Bollerslev, Tim/Chou, Ray Y.¡Kroner, Kenneth E (1992): ARCH Modeling in Finance: A Review of the Theory and Empirical Evidence, Journal of Econometrics, Vol. 52, S. 5-59.
Google Scholar -
Bollerslev, Tim./Wooldridge, Jeffrey M. (1992): Quasi-Maximum Likelihood Estimation and Inference in Dynamic Models with Time Varying Covariances, Econometric Reviews, Vol. 11, S. 143-172.
Google Scholar -
Bollerslev, Tim/Engle, Robert F./Nelson, Daniel B. (1994): ARCH Models, Handbook of Econometrics, Volume IV, Chapter 49, S. 2959-3038.
Google Scholar
Abstract
Predicting Interest Rate Volatility with Regime-Switching Models: An Empirical Analysis of the Euro-Deutschmark Money Market
This contribution analyses the usefulness of the Generalized Regime-Switching-(GRS)-Model proposed by Gray (1996a, 1996b) for modelling and forecasting interest rate volatility in the Euro-Deutschmark money market. The theoretical part of the contribution begins by introducing the GRS model. It turns out that many known models such as GARCH and Markov switching may be regarded as a restricted variant of the GRS model. An empirical comparison with the traditional approaches shows that the GRS model is the superior option for describing the dynamics of the interest rate volatility of both one-month and three-month money. Moreover, one-step forecasts suggest a good out-of-sample performance of the GRS model. Irrespective of the GRS model's complexity, its recursive representation allows it to be implemented easily.