fix: NormalDist.log_pdf divide SD by sqrt(weights) instead of weights#458
Open
DEEP-600 wants to merge 2 commits intodswah:mainfrom
Open
fix: NormalDist.log_pdf divide SD by sqrt(weights) instead of weights#458DEEP-600 wants to merge 2 commits intodswah:mainfrom
DEEP-600 wants to merge 2 commits intodswah:mainfrom
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
NormalDist.log_pdfwas dividing SD byweightsinstead ofsqrt(weights), making the effective variancescale² / w²instead ofscale² / w. This disagrees with the GLM convention used byV(),phi(), and_W()throughout the codebase.Change:
self.scale / weights→self.scale / np.sqrt(weights)(one line in distributions.py)Tests: Added test_distributions.py with 4 tests — weights=1, weights>1, vectorized, and None default. Full suite passes (153 passed, 0 failures).
Impact: Fixes log-likelihood and downstream stats (AIC, AICc, GCV, pseudo R²) for
LinearGAMwith non-uniform sample weights. Fitted coefficients are unaffected since PIRLS usesV()and_W()which were already correct.Fixes #457