Q2. What future works have the authors mentioned in the paper "Forecasting oil price realized volatility using information channels from other asset classes" ?
An interesting direction for further research would be the use of their forecasting strategy for the prediction of other assets ’ volatilities.
Q3. What is the way to forecast oil price volatility?
(ii) The HAR-RV-X models that combine asset volatilities from all asset classes are the best performing models, since they capture the different “information channels” that impact on oil price volatility at different times.
Q4. What is the main currency that affects oil prices?
As far as the foreign exchange variables are concerned, the authors maintain that theEUR/USD is the main currency that exercises an impact on oil fluctuations, while the use of the GBP/USD futures is incontestable, given that it is related to the Brent crude oil.
Q5. How do the authors evaluate oil volatility forecasts?
(i) The authors consider 14 exogenous variables (using HAR-RV-X models), which are categorized into four different asset classes (Stocks, Forex, Commodities and Macro) and the authors investigate whether their realized volatilities improve the oil volatility forecasts.
Q6. What is the definition of realized volatility?
Realized volatility is based on the idea of using the sum of squared intraday returns to generate more accurate daily volatility measures.
Q7. What is the forecasting accuracy of the models illustrated in Section 5?
The forecasting accuracy of the models illustrated in Section 5 is initiallyevaluated using two well established evaluation functions, namely the Mean Squared Predicted Error (MSE) and the Mean Absolute Predicted Error (MAE):∑ () , (15)and∑ || , (16)whereis the s-days-ahead oil realized volatility forecast, whereasis the Brent Crude oil realized volatility at time t+s.
Q8. What is the main reason why oil prices have fallen?
Even more, this fall in oil prices has resulted in increased oil price volatility, which is an essential input in many macroeconomic models, as well as, in option pricing and value at risk.
Q9. How does the HAR-RV-COMBINED model reduce the forecasting error?
the HAR-RV-COMBINED model reduces the forecasting error by more than 10% in the long-run horizons, compared to the single HAR-RV.
Q10. What do Bollerslev and Wright (2001) maintain that any volatility series exhibits?
Bollerslev and Wright (2001) maintain that any volatility series exhibits long-memory behaviour and so a model which considers this stylized fact (such as the HAR model) is able to produce better forecasts.