Fix dev_skewnorm() residual sign to use the mode, not the mean - #144
Open
nehill197 wants to merge 1 commit into
Open
Fix dev_skewnorm() residual sign to use the mode, not the mean#144nehill197 wants to merge 1 commit into
nehill197 wants to merge 1 commit into
Conversation
The deviance magnitude is computed relative to the fitted distribution's mode (mode_sat), since that is provably where the saturated log-likelihood is achieved for a location family. But the residual sign was computed relative to the theoretical mean instead. For skewed shapes, mode != mean, so points between the two got a small deviance (close to the mode) but the wrong sign (below the mean), producing a gap in the residual distribution where no small positive (or negative) residuals could occur. Use mode_sat for the sign too, matching the reference point used for the magnitude. dev_skewlnorm() inherits the fix since it delegates to dev_skewnorm().
nehill197
marked this pull request as draft
July 17, 2026 19:16
nehill197
marked this pull request as ready for review
July 17, 2026 20:24
joethorley
approved these changes
Jul 17, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The deviance magnitude is computed relative to the fitted distribution's mode (mode_sat), since that is provably where the saturated log-likelihood is achieved for a location family. But the residual sign was computed relative to the theoretical mean instead. For skewed shapes, mode != mean, so points between the two got a small deviance (close to the mode) but the wrong sign (below the mean), producing a gap in the residual distribution where no small positive (or negative) residuals could occur. Use mode_sat for the sign too, matching the reference point used for the magnitude.
dev_skewlnorm() inherits the fix since it delegates to dev_skewnorm().