Timeline for Loss functions with one-step-ahead volatility forecasts & volatility proxy
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 24, 2017 at 23:28 | vote | accept | Hsk | ||
| Jun 23, 2017 at 14:43 | answer | added | Richard Hardy | timeline score: 1 | |
| May 4, 2017 at 17:34 | history | edited | Hsk | CC BY-SA 3.0 |
added 1 character in body
|
| May 4, 2017 at 17:15 | comment | added | Richard Hardy | Regarding 2), just pretend the simulated data is the real data and act accordingly. I.e. use the outer product of the estimated errors (if the conditional mean model is "empty", then just the raw returns) $\hat\varepsilon_t \hat\varepsilon_t'$ where needed. | |
| May 4, 2017 at 16:56 | comment | added | Hsk | The paper is Becker et al (2015) -Selecting volatility forecasting models for portfolio allocation purposes. They haven't really given a reason for using that as a proxy, simply stated it. Regardless of whether it's a noisy proxy, if I was to follow what they did, can you please give me an answer to 2)? | |
| May 4, 2017 at 15:32 | comment | added | Richard Hardy | That will be a very noisy proxy. Does the study you are refering to give a good reason for using a proxy when they know the true value? Sounds like wasting statistical power. | |
| May 4, 2017 at 15:22 | comment | added | Hsk | Oh okay, now I understand what you meant, thanks for clarifying. Yes I have specified the DGP as DCC GARCH and simulated using the sd calculated by univariate GARCH. However, this paper that I am following has used the outer product of error terms as proxy for the loss function and since I am trying to replicate their study, that's why I want to use the proxy they have used. | |
| May 4, 2017 at 14:11 | comment | added | Richard Hardy | You know the true volatility, not a proxy. How do you simulate? You must specify the data generating process when you simulate (it is impossible to do without it). The volatility is part of the specification of the data generating process. Since you specify it, you know it. | |
| May 4, 2017 at 13:03 | comment | added | Hsk | Can you please elaborate how I can calculate the true underlying volatility? I'm afraid i am totally new to this so need clarification regarding the implementation. I would really appreciate if you could also tell me why I know the true underlying proxy with simulated returns | |
| May 4, 2017 at 13:02 | history | edited | Hsk | CC BY-SA 3.0 |
edited title
|
| May 4, 2017 at 13:01 | comment | added | Richard Hardy | With simulated returns, you know the true underlying volatility. Use it. Using some proxy instead would be inefficient. | |
| May 4, 2017 at 12:53 | history | edited | Hsk | CC BY-SA 3.0 |
added 59 characters in body
|
| May 4, 2017 at 12:47 | history | asked | Hsk | CC BY-SA 3.0 |