Q2. What are the future works mentioned in the paper "Long memory versus structural breaks in modeling and forecasting realized volatility" ?
Their short-memory-Break model is superior among most of the current forecasting methods if the future break dates and sizes are known.
Q3. How can the authors make the model-free realized volatility close to the underlying volatility?
By sampling intraday returns sufficiently frequently, the model-free realized volatility can be made arbitrarily close to underlying integrated volatility, the integral of instantaneous volatility over the interval of interest, which is a natural volatility measure.
Q4. What is the problem for measuring, modeling and forecasting conditional volatility?
An inherent problem for measuring, modeling and forecasting conditional volatility is that the volatility is unobservable or latent, which implies modeling must be indirect.
Q5. How can the authors get out-of-sample forecast performance?
In summary, even though the DGP is pure mean break series without any long memory,we still can get very good out-of-sample forecast performance using simple AR-I(d) model.
Q6. What is the way to forecast the volatility of the Yen?
Realized volatility constructed by intraday high-frequency data improves its out-of-sample forecasts ability compared with traditional volatility models.
Q7. What is the significance of the BP DGP?
The important implication from this Monte Carlo evidence is that the long memory DGP provides a good parsimonious alternative of in-sample fit for the true structural-break DGP when the authors have little knowledge for the past break dates and size.
Q8. Why does the former model have a relatively fast adaptability to the current volatility?
The main reason is because thatthe former model, which exploits the intraday volatility information, provides a relative accurate and fast-adapting estimate of current volatility while the latter model, depending on slowly decaying past squared returns, adapts only gradually to the current volatility shocks.
Q9. What is the spectral density of a time series process?
A time series process, ty , with autocorrelation function kρ at lag k, is a long memoryprocess whenlim n kn k n ρ →∞ =− → ∞∑ (5)The spectral density 2( ) ( / 2 ) ikkkf e ωω σ π ρ ∞ − =−∞ = ∑ tends to infinity at zero frequency,(0)f = ∞ .
Q10. What is the slow decay of the autocorrelations in Figure 1?
According to the slow decay of autocorrelations in Figure 1, it is evident that thelogarithmic realized volatility of the exchange rate series appears to have long memory dynamics.
Q11. What is the forecasting model for the real financial volatility series?
This result shows that long memory/fractional integrated model will still be the best forecasting model when the true financial volatility series are generated by structural breaks and the authors have little knowledge about these breaks information.
Q12. How do they compare the results of the ABDL models?
Although the Bayesian information criteria select a fourth-order VAR, the authors use afifth-order model to compare their result to those in ABDL.10 Forecasts are obtained by estimating rolling models.