Timeline for Accounting for uncertainty in evaluating a forecast
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 21, 2020 at 17:35 | comment | added | kmi | Thanks for explaining. That makes a lot more sense. | |
| Jul 20, 2020 at 14:56 | comment | added | Akylas Stratigakos | Assume that the data comes from a Normal distribution. Normal distribution is continuous and unbounded, so you cannot actually estimate 100% Prediction Interval. Also, you'll always have to work with a finite sample and use it to approximate the true underlying process. Estimating 100% from the data would correspond to the range of the data. | |
| Jul 16, 2020 at 22:01 | comment | added | kmi | So, I think my question is what makes 98 that different from 100? This is the question no one answers. | |
| Jul 3, 2020 at 12:51 | review | First posts | |||
| Jul 3, 2020 at 13:04 | |||||
| Jul 3, 2020 at 12:44 | history | answered | Akylas Stratigakos | CC BY-SA 4.0 |