Predicting returns

In the previous exercises, we have assumed a constant mean by setting

garchspec <- ugarchspec(mean.model = list(armaOrder = c(0,0)),
                        variance.model = list(model = "gjrGARCH"),
                        distribution.model = "sstd")

In practice the predicted return \(\mu_t\) is often time-varying. To capture this time-variation, one can use a GARCH-in-mean, AR(1), MA(1), ARMA(1,1) model, among others.

These models are implemented in rugarch and can be specified by changing the arguments for mean.model.

Which of the following statements is incorrect?

Answer the question

50 XP

Possible Answers

We specify a MA(1) model by setting mean.model = list(armaOrder = c(0,1)).
press1
We specify a GARCH-in-mean model by setting mean.model = list(armaOrder = c(0,0), archm = TRUE, archpow = 2).
press2
A GARCH-in-mean model exploits the presence of correlation between today's and tomorrow's return.
press3
GARCH models require to jointly specify the mean, variance and distribution models.
press4