Skip to content

Publish book-reconstruction evidence and correct partial-fill queue payoff - #16

Merged
Aroesler1 merged 1 commit into
mainfrom
review/lob-evidence-audit
Sep 10, 2026
Merged

Aroesler1 merged 1 commit into
mainfrom
review/lob-evidence-audit

Conversation

@Aroesler1

Copy link
Copy Markdown
Owner

Committed aggregate evidence and offline verifiers make the book, queue and simulator claims publicly checkable. Direct marked payoff now weights partial fills by executed shares and counts unavailable marks, preserving historical scores beside the repaired measure.

On fifteen selected symbol-days, front-queue direct payoff beats the back in 6 sessions and is negative in 14. These are descriptive marked-inventory results before liquidation and fees; they do not establish a causal queue-position benefit. Both simulators remain distinguishable from real data.

Validation: full offline Python suite and committed-result verifiers pass locally. No fresh WRDS connection or model call was used for finalization. Review branch only; no main merge.

@Aroesler1
Aroesler1 merged commit caad8de into main Sep 10, 2026
4 checks passed
@Aroesler1
Aroesler1 deleted the review/lob-evidence-audit branch September 10, 2026 09:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant