Skip to main content
12 events
when toggle format what by license comment
Nov 10, 2023 at 13:29 comment added Robert Long @RichardHardy thanks again, I have modified that sentence as per your suggestion. Cheers :)
Nov 10, 2023 at 13:29 history edited Robert Long CC BY-SA 4.0
Clarity
Nov 10, 2023 at 12:51 comment added Richard Hardy Thank you. I think your answer could be even more helpful if you adjusted the sentence Differencing is applied when your time series displays a trend or seasonality, as it seems to have substantial potential to mislead.
Nov 10, 2023 at 11:44 history edited Robert Long CC BY-SA 4.0
To address the comment by Richard Hardy
Nov 10, 2023 at 11:28 comment added Robert Long @RichardHardy I agree with you. I will update my answer accordingly.
Nov 10, 2023 at 11:12 comment added Richard Hardy Differencing is applied when your time series displays a trend or seasonality. This kind of treatment leads all too many practitioners to overdifferene their time series, introducing a moving-average unit-root component and increasing the error variance. Existence of deterministic trend or seasonality without presence of a unit root (seasonal or not) does not warrant differencing.
Nov 10, 2023 at 11:06 comment added Marco Bresson I did, but my vote was not casted due to my low reputation. I casted it again and it went through this time around!
Nov 10, 2023 at 10:38 comment added Robert Long You're welcome. Glad to help. Thanks for accepting the answer. Please consider giving my answer your upvote too :)
Nov 10, 2023 at 10:34 vote accept Marco Bresson
Nov 10, 2023 at 10:34 comment added Marco Bresson Thank you so much for your detailed answer! You made this very clear. Have a good day
Nov 10, 2023 at 10:27 history edited Robert Long CC BY-SA 4.0
Clarity
Nov 10, 2023 at 10:00 history answered Robert Long CC BY-SA 4.0