feat(forecasting): read-only OOS backtest of §9.6 rate-sensitivity (#978) #1024

Merged
bot-backend merged 1 commit from feat/951b-backtest-rate-sensitivity into main 2026-06-03 09:33:02 +00:00
Collaborator

EPIC 7 валидация — таск #978: read-only OOS-бэктест §9.6. scripts/backtest_rate_sensitivity.py (read-only, Source B, time-ordered holdout, point-in-time OOS hit-rate + honesty-block + per-tier + verdict). 36 тестов.

⚠️ ВЕРДИКТ (прогон vs прод 2026-06-03): EKB-wide OOS hit-rate 0.148 (< 0.5 coin-flip), лаг нестабилен → §9.6-движок НЕ валиден OOS → остаётся ADVISORY. Вероятно confounded survivorship-bias'ом Source B (инфляция свежих месяцев → ложный тренд). Чистая ре-валидация — на Source A (corp_room_month, без survivorship; сейчас 13 мес, добирает). Refs #978 #951

**EPIC 7 валидация — таск #978: read-only OOS-бэктест §9.6.** `scripts/backtest_rate_sensitivity.py` (read-only, Source B, time-ordered holdout, point-in-time OOS hit-rate + honesty-block + per-tier + verdict). 36 тестов. ⚠️ **ВЕРДИКТ (прогон vs прод 2026-06-03):** EKB-wide OOS hit-rate **0.148** (< 0.5 coin-flip), лаг нестабилен → §9.6-движок **НЕ валиден OOS → остаётся ADVISORY**. Вероятно confounded survivorship-bias'ом Source B (инфляция свежих месяцев → ложный тренд). Чистая ре-валидация — на Source A (corp_room_month, без survivorship; сейчас 13 мес, добирает). Refs #978 #951
bot-backend added 1 commit 2026-06-03 09:33:01 +00:00
Add scripts/backtest_rate_sensitivity.py — STRICTLY READ-ONLY out-of-sample
validation gate for the §9.6 engine (#1009). Monthly sold-units from Source B
(objective_lots) EKB-wide + per class, aligned to monthly key_rate, time-ordered
holdout (fit oldest 70%, test newest), point-in-time OOS directional hit-rate +
in-sample-vs-OOS honesty block + per-tier lift + coin-flip+lag-stability verdict.
Deterministic, no LLM, no DB writes. CLI --since/--holdout-frac/--classes/--json.
36 unit tests on synthetic.

VERDICT (run vs prod 2026-06-03): EKB-wide OOS hit-rate 0.148 (< 0.5 coin-flip),
lag unstable → §9.6 engine NOT OOS-validated → stays ADVISORY. Likely confounded
by Source B survivorship bias (recent-month inflation → spurious trend); clean
re-validation needs Source A depth (13mo now, deepens weekly) or survivorship
correction. The harness re-validates when data improves.
bot-backend merged commit 6f06ca7f16 into main 2026-06-03 09:33:02 +00:00
bot-backend deleted branch feat/951b-backtest-rate-sensitivity 2026-06-03 09:33:02 +00:00
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: lekss361/gendesign#1024
No description provided.