Testing the Presence of Structural Break in the Euro Exchange Rate Variance

Abstract

We investigate the presence of structural breaks in the euro exchange rate conditional variance from the GARCH models and how does it influence the Value at Risk metrics. We use daily exchange rate for the euro/dollar, euro/£, euro/CHF, euro/yen for the 01:1999 - 09:2013 in order to test the influence of structural breaks on the GARCH models. The structural break in the series mean and variance are identified using the PELT algorithm, the structural breaks dates are captured using dummy variables in the GARCH models, the selection of models is done using the informational criterion [Akaike, Schwarz, Log-likelihood]. We find evidence of structural breaks in the unconditional variance of euro exchange rate return series over the 1999-2013 period, we test the performance of the structural GARCH(1,1) models versus the GARCH(1,1) and find that structural GARCH models outperforms when forecasting exchange rate return volatility in real time.