papers: honesty + citation follow-up (paper-1 supersession, reverse-import, 2022 priority) - #53
Merged
Merged
Conversation
…mport, 2022 priority) Paper 1: dated supersession note retiring the Sections 4-5 likelihood-vs-Schur SELECTION contrast (objective-mismatch artifact, not a likelihood failure; the held-out likelihood does not pick ill-conditioned estimates); flags Section 6 'Schur-LW' as cross-block ENTRY shrinkage, not the structured Schur damping. Two Sides of Schur Damping: separate the (classical) shrinkage intensity from the Schur OPERATION (cite Schaefer & Strimmer 2005); add the reverse-import section (a ridge/robust hedge repairs the structured Schur estimator, shipped as SchurConditionalCovariance, best-on-likelihood); establish the 2022 priority of the allocation gamma-dial (Geek Culture, Nov 21 2022). refs.bib: correct cotton2024 title; add schafer2005, cotton2022. Regenerate docs web edition (pandoc 3.9.0.2). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…p-to-date) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to #50, applying review findings.
Paper 1 (Schur Pseudo-Likelihood): dated supersession note retiring the §4–5 likelihood-vs-Schur selection contrast — a controlled 2×2 shows the poor showing is objective mismatch (judged by portfolio variance, not likelihood), not a failure of the likelihood, which also does not select ill-conditioned estimates. Flags §6 "Schur–LW" as cross-block entry shrinkage, not the structured Schur damping it's named for.
Two Sides of Schur Damping: separates the (classical) shrinkage intensity from the Schur operation (cites Schäfer & Strimmer 2005); adds the reverse-import section — a ridge/robust hedge repairs the structured Schur estimator (best-on-likelihood), shipped as
SchurConditionalCovariance; establishes the 2022 priority of the allocation γ-dial (Geek Culture, Nov 21 2022).refs.bib: correct
cotton2024title; addschafer2005,cotton2022. Docs web edition regenerated (pandoc 3.9.0.2).Note: references
SchurConditionalCovariance(#52) — merge #52 first.🤖 Generated with Claude Code