Timeline for GARCH modeling - sliding or expanding window?
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 15, 2018 at 17:35 | comment | added | phdstudent | That is fair enough. | |
| Mar 15, 2018 at 17:34 | comment | added | John | I tend to prefer more data to less. If there are regime shifts or other features that a simple GARCH model cannot handle, then there's a problem with model specification. Fit a different model. | |
| Mar 15, 2018 at 16:18 | comment | added | phdstudent | If volatility is time-varying, that is the reason why sliding windows are preferable for out-of-sample predictions. You want to take those regime shifts into account when you run the model. | |
| Mar 12, 2018 at 16:46 | history | answered | John | CC BY-SA 3.0 |