Repository navigation
Description of Underlining Statistic Methods Used #100
Description
I think it would be nice to have some descriptions about how the statistics of analysis.py works (compute_change_points) and the math behind it. This can be helpful for those investigating to use this as possible solution for various use-cases. Short Example:
analysis.py is a smart change point detector built by combining:
- A powerful segmentation algorithm (EDivisive)
- A fast statistical test (t-test)
- Sliding windows for scalability
- A postprocessing step to enforce quality
- Optional incremental processing
It produces stable, noise-resistant change points for industrial or scientific signals.
Metadata
Metadata
Assignees
Labels
No labels
Activity
henrikingo commentedon Nov 28, 2025
cc @Sowiks
Sowiks commentedon Dec 6, 2025
I will work on this next. Thank you!
Sowiks commentedon Feb 9, 2026
Please find the pull request here: #126
ligurio commentedon Mar 4, 2026
@Gerrrr seems done, #126 is merged
Gerrrr commentedon Mar 4, 2026
Yes, thanks for the reminder @ligurio! FWIW I will also upload this change to the website today-ish.
Gerrrr commentedon Mar 6, 2026
Published to the website - https://otava.apache.org/docs/math