Return period calculation based on a Gumbel-distribution fit to sample data #29
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The plan is to exclusively rely on scipy for the functionality in the future, but for now to provide a substitute that doesn't rely on a version of scipy that is still under development (as of Dec 2024). Co-authored-by: Christopher Polster <[email protected]>
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #29 +/- ##
===========================================
+ Coverage 98.68% 98.71% +0.03%
===========================================
Files 8 8
Lines 684 703 +19
Branches 26 26
===========================================
+ Hits 675 694 +19
Misses 7 7
Partials 2 2 ☔ View full report in Codecov by Sentry. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds distribution and extreme_values submodules to earthkit.plots.stats for computing recurrence statistics based on a Gumbel distribution fit to a data sample. Can be used, e.g., to compute return periods of extreme precipitation of flooding events in a given time interval.
The implementation is class-based, since the fitted distribution parameters need to be stored for subsequent invocations of the statistics calculations. The interface for the distributions so far matches that of scipy's rv_continuous and the distributions are only reimplemented here to allow for fitting of multiple distributions in a vectorised fashion along a given axis (scipy only fits to 1D samples).
The Gumbel fit makes use of scipy's lmoment function, which was only added in the latest 1.15 release, so a vectorized implementation by @corentincarton is included for environments with older versions of scipy. The included implementation only works along the first array axis, but that covers the common use-case of fitting a distribution to a (time)series of fields. The scipy implementation supports application along any axis and is therefore preferred, despite being a little slower.