A better model for EUR/USD returns
In the previous exercise you have analyzed the statistical significance of the estimated parameters of the AR(1) GJR GARCH model with skewed student t distribution for the daily EUR/USD returns. The conclusion is that we should simplify the GARCH model used. Let's therefore take a constant mean standard GARCH model with student t distribution. We fix the mean value to zero and use variance targeting.
Este ejercicio forma parte del curso
GARCH Models in R
Instrucciones del ejercicio
- Complete the code to estimate a constant mean standard GARCH model with student t distribution and variance targeting.
- Use
setfixed()
to impose that the mean parameter equals 0. - Estimate the model.
- Check visually that these changes lead to a volatility series close to the one using
flexgarchfit
.
Ejercicio interactivo práctico
Prueba este ejercicio completando el código de muestra.
# Specify model with constant mean, standard GARCH and student t
tgarchspec <- ugarchspec(mean.model = list(armaOrder = ___),
variance.model = list(model = ___, variance.targeting = ___),
distribution.model = ___)
# Fix the mu parameter at zero
___(tgarchspec) <- list("mu" = 0)
# Estimate the model
tgarchfit <- ___(data = EURUSDret, spec = tgarchspec)
# Verify that the differences in volatility are small
plot(sigma(___) - ___(flexgarchfit))