Repository navigation
Add strategy-selection bias research - #5
Merged
Merged
Conversation
asoracca
marked this pull request as ready for review
September 22, 2026 15:43
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a reproducible zero-alpha strategy-selection experiment, development-only CSCV/PBO diagnostics, confidence intervals, recorded results and offline CI coverage.
Selection and CSCV use only the first 120 of 240 monthly observations. The selected strategy is frozen before evaluation on the final 120 months. A deliberately biased hindsight winner is measured on a separate synthetic dataset using an independent seeded random stream.
The branch includes current main and preserves the accounting/storage tests and CI commands from PR #6. Added regression tests verify that changing evaluation returns cannot change the selected strategy or development diagnostic.
Validation: 46 local tests passed, lint passed, accounting golden demo and SQLite replay passed, and the seeded study and benchmark smoke run completed. Updated chart, summary CSV and manifest record actual results, configuration, versions and checksums.
Scope: synthetic methodology demonstration, not evidence of historical or live trading alpha. Full candidate histories are required for meaningful selection-bias analysis.