spb/anomaly-atlas Public License

1---2project: anomaly-atlas3document: expA_data_reality/analysis4author: Simon-Pierre Boucher5contact: contact@spboucher.ai6data_source: hfmarketdata.io7created: 2026-08-128modified: 2026-08-129status: reviewed10---1112# Analysis — expA_data_reality1314Run: `results/expA_data_reality/20260812T054515Z/results.json`15(hardware manifest embedded; 54 network requests on first pass, 100 % cache16hits on re-run — the reproducibility anchor works).1718## What was established19201. **Granularity is real.** 1-minute bars are served for every advertised21   class; 1min is the floor. Options are daily chain snapshots (17 columns,22   with bid/ask/IV/Greeks/OI) over 67 quarters, 2010_q1 → 2026_q3_partial.232. **Depth.** stock/etf 1min from Jan 2000 (26+ years); futures & index from24   Jan 2008; fx from 2010; crypto from Apr 2013. Archive lags a few days25   (equities ended previous Friday on run date) — not a live feed.263. **Timestamps** are US/Eastern wall-clock, bar-start labeled, no TZ marker.27   Sessions: equities 04:00–19:59 ET, fx Sun 17:00 → Fri 16:59, futures ≈24 h,28   crypto 24/7, SPX prints until 16:20. Cross-asset closes are therefore29   non-synchronous **by construction**.304. **Bars exist only where trades occurred** — no zero-volume placeholders.31   Liquid RTH is complete (390/390); an illiquid name printed 38 bars/day.325. **Daily ≠ aggregated 1min.** Daily bars are RTH-only with the official33   auction close (312.41 vs 312.49) and include auction/consolidated volume34   missing from 1min bars (46.1 M vs 34.7 M extended / 25.7 M RTH).356. **Adjustments verified** on the AAPL 4:1 split (exact), and `adj_splitdiv`36   re-bases the whole history to the build date — adjusted series are not37   point-in-time stable.387. **Limits.** Hard 50 000-row JSON cap (pagination verified: 218 006 rows of39   SPY 1min/2020, 0 dupes, monotonic); no rate limiting observed; 0.08–0.22 s40   small-request latency.4142## Interpretation (confidence framing)4344No anomaly is claimed; Level n/a. The experiment succeeded by its own45criterion: every downstream-critical property was pinned down. Items 3–6 are46direct seeds for the artifact taxonomy — each is a mechanism that can47manufacture a fake anomaly (spurious lead-lag from non-synchronous sessions;48spurious autocorrelation from stale prints; phantom overnight returns from49auction-vs-last-bar close mismatch; look-ahead from rolling adjustment).5051## Verdict5253**Complete.** Proceed to expB_artifact_baselines (measure the artifact nulls);54in parallel, Phase 1 literature work can start. Everything downstream must55join bars explicitly on time with declared staleness rules — never implicit56LOCF — and must state which close (auction vs last-bar) it uses.57