Closed davharris closed 6 years ago
Observer models are already discussed prior to this, so I see no problem putting those details in this paragraph. Otherwise it's definitely an improvement. It might also help to include the ARIMA(0,1,0) notation for the naive model.
Okay, I've added more details and a line in the reviewer response. Might read awkwardly now, but it's probably good to have all this information available. Thanks @sdtaylor for your input above.
LGTM
Thanks both! Fixed the grammar, and merging now that it's been approved by both of you.
If we like this, I can add a comment for the reviewers.
Ideally, we should also discuss where the standard deviations of ε come from, but that differs for the observer-corrected and non-corrected "average" models so I can't figure out the best way to discuss it without breaking the flow.