SPB Git forge
spboucher.ai ↗

spb/anomaly-atlas

Public License ★ Pinned

Systematic discovery & rigorous validation of statistical anomalies in open HF market data (hfmarketdata.io) — pre-registered, artifact-null-driven, fully reproducible. Live atlas: www.anomaly-atlas.io

quant-researchmarket-microstructureanomaliesreproducibilityduckdbapple-silicon
18commits 1branches 0releases
396.0 KBsize
maindefault branch
21 days agolast push
Python 61.4% JavaScript 28.7% CSS 8.6% Shell 0.7% Makefile 0.5%

History of research/artifact_taxonomy.md · clear filter

  1. Wed, Aug 12, 2026 5

    1. fix: expE lint (semicolons) + duplicated Next-experiment block
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      Simon-Pierre Boucher committed 1 mo ago · 1 file +11 −4
    2. expD: lead-lag scan — decay on synchronized pairs, splice-invariant ES-SPY, SPX content staleness
      …
      - 2006-07: real both-fresh diffusion (SPY->sectors +0.07..+0.148 FDR);
  2014-15: collapse to tiny residuals (a few FDR survivors -> expF;
  H08 expected-negative clause formally falsified, recorded)
- ES->SPY -0.032 both-fresh, identical across all 3 splice variants
- SPX->SPY +0.132 both-fresh (fresh_frac .996): content staleness beats
  print synchronization — H11 confirmed at scan level
- T3 amended: LOCF artifact non-monotone in staleness (inflates mid-stale,
  dilutes ultra-stale)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      Simon-Pierre Boucher committed 1 mo ago · 1 file +7
    3. expC: reversion scan — null mis-specification caught, corrected triage
      …
      - pre-specified EDGE+Roll intraday bounce null predicts impossible AC1
  (-3..-27): discarded its positive-excess family as null artifact;
  variance-consistency rule adopted (taxonomy T1 amended)
- corrected MA(1)-consistent VR triage: 14 Level-0 cells (JPM 1min -0.243,
  MSFT 5min 2000-07...) + daily 2008-15 mega-cap/index reversal family
  (XOM -0.133 FDR, SPY -0.055 FDR)
- H01 confirmed (liquid intraday lag-1 reversion = 0 net of bounce);
  decay clause partially falsified — recorded as-is
- new: stats/multiple_testing.py (BH-FDR, bootstrap p), EDGE wrapper,
  pre-specified splits in data/universe.py, 6 new gate tests

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      Simon-Pierre Boucher committed 1 mo ago · 1 file +7
    4. expB: artifact baselines measured (after §8.1 synthetic gate)
      …
      - synthetic generators with known ground truth + 11-test gate; the gate
  caught a real VR estimator bug (double division by q) before real data
- stats: reversion (VR/AC1/half-life), leadlag, block bootstrap;
  validation: Roll spread, excess reversion, staleness, LOCF
- expB (pre-specified protocol): 31 tickers Q1 2024 RTH 1min —
  bounce AC1 -0.009/-0.051/-0.232 by staleness tercile, VR30 down to 0.55
  with zero economics, SPY spuriously leads stale names (+0.047 at +1min,
  Spearman +0.43), SPX lags SPY by 1min (+0.065)
- artifact_taxonomy.md: 7 entries T1-T7 with measured magnitudes

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      Simon-Pierre Boucher committed 1 mo ago · 1 file +93 −5
    5. Bootstrap: full research skeleton (CLAUDE.md §3, localvm-research layout) + web platform
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      Simon-Pierre Boucher committed 1 mo ago · 1 file +17