We can take the variance series and apply ARIMA model on it to have forecasting of volatility.
Someone answered it here as:
"ARIMA modelling is not the best in this circumstance because it models the mean rather than the variance. Therefore ARCH modelling is preferable"
What does this mean?
For several reasons, including that the number of observations must be greater than n>=50
This is simply because the variables in an ARIMA model are the variable of interest and the residuals; i.e. the time-dependence structure of the variance is not represented in an ARIMA model. The ARIMA model is usually called the mean equation.
This is why we need to have a separate model (GARCH) to represent the time-dependence structure of the variance, i.e. the variance equation.
- Badges
- Science topic
- Similar topics
- Correction