dennisthiessen
9704e0d85a
Add research-only snapshot extender, PIT dollar-volume mask for signal IC, rank-only harness path, fingerprint+breadth runner, and docs. Fingerprint reproduced IC -0.045 / t -2.91 on prod.sqlite. No production gate/schedule changes.
1.8 KiB
1.8 KiB
Broad-universe fip_id IC research (Phase B)
Generated: 2026-07-18T19:48:28.127710
Scope
- Research only — production universe, gate, scanner, schedule unchanged.
- Price-only signal harness; no sentiment/fundamentals on the broad tier.
- Point-in-time liquidity mask: top 1500 by 63d median $vol, price ≥ $5.0 at as-of.
Caveats
- Survivorship bias: today's constituents backfilled historically (worse in small caps).
- IEX volume undercount: relative $vol rank only, not absolute floors.
- Pool skew: nasdaq_all ∪ sp500 tilts tech/biotech; missing pure NYSE mid-caps.
Fingerprint (505-name prod snapshot)
- Expected: IC ≈ -0.045, t ≈ -2.9
- Observed: IC = -0.045, t = -2.91, weeks = 35, reliable = True
- Pass: True
Liquid-breadth signal_eval (fip_id)
Not run yet in this environment (no Alpaca credentials to build
backtest_snapshots/research.sqlite). Local machine with keys:
python scripts/extend_snapshot_universe.py --force-copy
python scripts/run_fip_breadth_research.py --skip-fingerprint --allow-spawn --workers 6
(--skip-fingerprint only after a green fingerprint on this machine.)
| metric | value |
|---|---|
| mean_ic | pending |
| ic_t_stat | pending |
| weeks | pending |
| avg_cross_section | pending |
| reliable | pending |
Verdict (iron rule)
- Fingerprint: green (IC −0.045 / t −2.91 / reliable / 35 weeks).
- Breadth iron rule: pending until research.sqlite run completes.
- Green breadth would authorize a follow-up proposal only (two-tier universe / gate revalidation) — not production wire-in.
Artifacts
- Fingerprint report:
reports/fip-breadth-20260718-194828-fingerprint.json(and summaryreports/fip-breadth-20260718-194828.json) - Breadth report: pending