Phase 2: implement descriptive clustered estimators and percentile intervals #663

Merged
erikinkinen merged 6 commits from phase2/p2-14-c-descriptive-statistics into main 2026-10-27 01:44:38 +01:00
Owner

Task

Closes #118 (P2-14-C).

Summary

Implement declared descriptive means, pooled ratios, linear quantiles and
complete paired seed/cell contrasts. Preserve operands, cell effects and
missing reasons. Recompute each estimator in all 99,999 complete-cluster
resamples and retain undefined draws without replacement.

Ordinary 95% percentile intervals are suppressed when more than 5% of draws
are undefined: 4,999 is permitted and 5,000 is suppressed. These intervals
are descriptive; confirmatory bounds and error allocation remain separate.

Validation

Original source-bound locked Python 3.14.8/NumPy 2.5.4 and portable fixtures
pass seven analytical groups and 32 negative cases. All 99,999 nonlinear
ratios match an exact-rational reference; another 99,999 paired estimates
retain unequal dependent windows and their fixed effect. Ties, zeros and
the exact undefined-draw boundary are covered.

The original source and integration receipts remain under
artifacts/backlog/2.2.1/P2-14-C/. The integrated inventory contains 295 native
tests. This refresh includes the accepted #660 predecessor at
8ed9e598d030c8cc9b8d871e7d425e9628d70965. Implementation and earlier validation
records are unchanged; the final metadata adds the reconciliation receipt.
All eight local hooks and the frozen specification scope passed.

Reconciliation evidence
has SHA-256 ccd054e09e3bed4503a8eb9dc45aa2c6b488ebeae6d22f3cc6cd49a9d0480734.

Acceptance gate and scope

Require all ten checks on this final commit, including both formal checks, and
closed native prerequisites before updating #118's single checklist and merging.
The fixtures establish method implementation; no campaign or scientific result
is supplied here. Complete csaI includes the requested Phase 5 supplement.

# Task Closes #118 (P2-14-C). ## Summary Implement declared descriptive means, pooled ratios, linear quantiles and complete paired seed/cell contrasts. Preserve operands, cell effects and missing reasons. Recompute each estimator in all 99,999 complete-cluster resamples and retain undefined draws without replacement. Ordinary 95% percentile intervals are suppressed when more than 5% of draws are undefined: 4,999 is permitted and 5,000 is suppressed. These intervals are descriptive; confirmatory bounds and error allocation remain separate. ## Validation Original source-bound locked Python 3.14.8/NumPy 2.5.4 and portable fixtures pass seven analytical groups and 32 negative cases. All 99,999 nonlinear ratios match an exact-rational reference; another 99,999 paired estimates retain unequal dependent windows and their fixed effect. Ties, zeros and the exact undefined-draw boundary are covered. The original source and integration receipts remain under `artifacts/backlog/2.2.1/P2-14-C/`. The integrated inventory contains 295 native tests. This refresh includes the accepted #660 predecessor at `8ed9e598d030c8cc9b8d871e7d425e9628d70965`. Implementation and earlier validation records are unchanged; the final metadata adds the reconciliation receipt. All eight local hooks and the frozen specification scope passed. [Reconciliation evidence](https://git.erikinkinen.fi/attachments/7f485c61-f68a-4d4e-8cdb-3622e1ef2fb6) has SHA-256 `ccd054e09e3bed4503a8eb9dc45aa2c6b488ebeae6d22f3cc6cd49a9d0480734`. ## Acceptance gate and scope Require all ten checks on this final commit, including both formal checks, and closed native prerequisites before updating #118's single checklist and merging. The fixtures establish method implementation; no campaign or scientific result is supplied here. Complete csaI includes the requested Phase 5 supplement.
erikinkinen added this to the Phase 2 milestone 2026-10-27 00:44:40 +01:00
Phase 2: record unchanged descriptive statistics after CI reconciliation (#118)
Some checks failed
clang-format / check-format (pull_request) Successful in 1m16s
markdownlint / markdown-lint (pull_request) Successful in 56s
formal / definitions (pull_request) Successful in 2m32s
ci / smoke (pull_request) Successful in 19m36s
ci / sanitizers (pull_request) Successful in 54m38s
markdownlint / markdown-lint (push) Successful in 1m7s
clang-format / check-format (push) Successful in 1m52s
formal / definitions (push) Successful in 2m50s
ci / smoke (push) Successful in 23m55s
ci / sanitizers (push) Has been cancelled
b25aa19d15
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
erikinkinen/AES!663
No description provided.