Timeline for Forecasting Multiple steps of a Multivariate Time Series for ALL Features
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 20 at 18:20 | history | edited | Arnav Tapadia |
edited tags
|
|
| Nov 20 at 18:19 | comment | added | Arnav Tapadia | Hi Stephan. Thank you for your comment. I hadn't know about compositional-data so that is really useful! Is it possible to do multistep forecasting with VAR non-recursively? I haven't been able to find a paper that does this and when I tried recursive forecasting with LSTMs, the results aren't the best over long forecast horizons, which is why I am worried about doing the same with another model – I plan to try it out though! | |
| Nov 20 at 7:09 | comment | added | Stephan Kolassa | If you are amenable to classical methods, vector autoregression (VAR) might be interesting. If I understand correctly, you have compositional-data, so you might include this term in your search. (I do wonder how you get your "emotional composition" at a resolution of a tenth of a second...) | |
| Nov 20 at 3:39 | history | edited | Arnav Tapadia | CC BY-SA 4.0 |
deleted 51 characters in body
|
| Nov 20 at 2:56 | comment | added | CommunityBot | Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. | |
| Nov 20 at 2:43 | history | edited | Arnav Tapadia | CC BY-SA 4.0 |
deleted 27 characters in body
|
| S Nov 20 at 2:41 | review | First questions | |||
| Nov 20 at 2:56 | |||||
| S Nov 20 at 2:41 | history | asked | Arnav Tapadia | CC BY-SA 4.0 |