Evaluating VaR (Value-at-Risk)

With this book we aim to contribute to the vast literature on conditional volatility models. Using the ARCH/GARCH class of models introduced in Engel¿s (1982) seminal paper we forecast one day ahead and ten days ahead Value-at-Risk on several exchange rates. The forecasts are done on a more volatile period than that period from which we estimate the models. We specify three models, GARCH(1,1), EGARCH(1,1) and GJR-GARCH(1,1) and test the models with three assumptions of the error distribution, normal, t and GED. We evaluate the models with Kupiec's (1995) test for unconditional coverage. The data ranges from January 1st 2006 through June 30th 2011. The results suggest that the GARCH(1,1) and GJR-GARCH(1,1) with normally distributed innovations are models adequately capturing the conditional variance in the series.