Estimation of GJR garch model
Just like any GARCH model, the GJR GARCH model is used to predict volatility. We use this model now to predict the volatility of the daily returns of Microsoft over the period 1999 till 2017.
These returns are available in the console as the variable msftret
. We have already computed the standard GARCH volatility predictions for you. They are available in the object sgarchvol
.
This exercise is part of the course
GARCH Models in R
Exercise instructions
- Specify the GJR garch model with a skewed student t distribution.
- Estimate the model.
- Compare the GJR GARCH volatility with
sgarchvol
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Specify the GJR GARCH model
garchspec <- ___(mean.model = list(armaOrder = c(0,0)),
___ = list(model = ___),
___ = ___)
# Estimate the model and compute volatility
gjrgarchfit <- ___(data = ___, spec = ___)
gjrgarchvol <- ___(___)
# Compare volatility
plotvol <- plot(abs(msftret), col = "grey")
plotvol <- addSeries(___, col = "red", on=1)
plotvol <- addSeries(sgarchvol, col = "blue", on=1)
plotvol