Timeline for How many data points to acurately approximate the average of recent values in a time series?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 31, 2017 at 21:49 | history | bounty ended | Phil | ||
| Jul 26, 2017 at 21:26 | comment | added | Carl | @eric_kernfeld Sure, why not? Another would be physically motivated models, which are highly dependent upon the physical circumstances, and, no specifics were given in the question. My point merely was that in general, modelling is needed to figure out what will work, and what will not. | |
| Jul 26, 2017 at 15:12 | comment | added | eric_kernfeld | ARIMA is useful as far as an error correlation model. It may be worth mentioning useful model classes for the function $T(t)$ itself -- for example, penalized splines. | |
| Jul 25, 2017 at 0:22 | history | edited | Carl | CC BY-SA 3.0 |
added 9 characters in body
|
| Jul 24, 2017 at 22:24 | history | answered | Carl | CC BY-SA 3.0 |