Repository navigation
Wind speed as a distribution: QQ quantiles, hybrid Weibull, exceedance, bootstrap over runs - #10
Merged
Merged
Conversation
distribution.compare: on the common sample, by region and lead group, forecast and observed quantiles at 1-99.5 % (QQ), the hybrid Weibull (calm fraction and a maximum-likelihood k and c for the rest; Takle and Brown 1978) and exceedance of 5/10/15 m/s, with a bootstrap over whole runs (model and observation paired) for each model-minus-observed quantile, k, c and exceedance ratio. Values are binned at 0.05 m/s per run, so a resample is a sum of histograms. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNh8pbXM7FCZppdye7QTLx
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNh8pbXM7FCZppdye7QTLx
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
RMSE and bias reward a forecast that hedges towards the middle; wind speed is skewed and bounded at zero, so
distribution.compare(pairs, "ws10")compares it as a distribution, on the common sample, by region and lead group (days 1–2, 3–5, 6–10):scipy.stats.weibull_min.fiton the same values);A bootstrap over whole runs, model and observation paired, gives 95 % intervals for each model-minus-observed quantile, k, c and the exceedance ratio. Values are binned at 0.05 m/s per run, so a resample is a sum of histograms.
Tests: the fit against SciPy, binned quantiles against exact ones, and a synthetic comparison. A model a fifth too slow is flagged in its 90th percentile, its scale and its 10 m/s frequency; an independent draw from the observed distribution is not. Plus the minimum-runs rule.
🤖 Generated with Claude Code
https://claude.ai/code/session_01XNh8pbXM7FCZppdye7QTLx