Forecast combination within and across models #127
nickcox896
started this conversation in
General
Replies: 1 comment 1 reply
-
|
Either is fine. Any combinations tend to improve forecast accuracy, but you generally get better improvements if the forecasts are not too similar. See https://robjhyndman.com/publications/combinations/ for the theory |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Is it common approach to take a combination within the same class of models or it should be across different models.
For example can I take the combination of various ARIMA models if this combination leads to better RMSES?
Beta Was this translation helpful? Give feedback.
All reactions