Timeline for Forecasting with BSTS and external predictors
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 30, 2022 at 21:32 | comment | added | user177196 | Well, identifying those insignificant one does not even require BSTS though. Or two different methods (e.g. regression vs BSTS) would give different sets of insignificant parameters, assuming we keep the same target variables? | |
| Jul 26, 2022 at 16:42 | comment | added | IrishStat | Stepdown is when you delete insigificant parameters from a model and just retain the statistically significant ones. | |
| Jul 25, 2022 at 21:21 | comment | added | user177196 | I am not sure I followed now. What do you mean with a stepdown? Even with such parameters, I am concerned about prediction accuracy vs exponential smoothing. | |
| Jul 25, 2022 at 20:50 | comment | added | IrishStat | or preferably do a stepdown and only use the statistically significant parameters | |
| Jul 25, 2022 at 13:25 | comment | added | user177196 | That comes with the caveat that we would need to fill in all the parameters you mentioned above, and many more, correctly? Which is a tough task on its own. | |
| Jul 25, 2022 at 11:26 | comment | added | IrishStat | I would think so as it is more general . | |
| Jul 25, 2022 at 4:08 | comment | added | user177196 | So in your opinion, does bsts provide a better time series forecast (both in terms of accuracy and robustness) versus exponential smoothing? | |
| Nov 3, 2017 at 10:01 | history | answered | IrishStat | CC BY-SA 3.0 |