โ€บNavigation
โš  VERDICT UNDER REVIEW IAAFT-dependent finding

Original verdict relied on IAAFT null calibration โ€” now prohibited by project policy. The finding needs to be re-derived from an approved null-calibration method before it can be used for decisions. Numbers below are preserved for the record.

Experiment 5 of 5 ยท panel-wide ยท 69 sec compute
STRONG SIGNAL 100% FLAGGED 69% PEARSON-BLIND

๐Ÿ”ฌ Were the 8 hand-picked pairs cherry-picked?

Verdict: STRONG signal โ€” no, they were not. When we ran Chatterjee on every one of 171 microstructure feature pairs on BTCUSDT and applied BH-FDR multiplicity correction at q=0.10, 100% of pairs are flagged, and 69% (118 of 171) are Spearman-independent (|ฯ| < 0.30) yet Chatterjee-decisive. The 8-pair pattern is the panel's dominant pattern.

Hub trail: audit hub โ†’ chatterjee-broadening/

Why this test mattered

We picked 8 feature pairs to test based on the prior audit's classifications. A skeptic could say: "Of course you found something โ€” you picked pairs that look interesting." This test eliminates that skepticism by running the same method on all 171 pairs we have access to, with multiple-comparison correction (BH-FDR at q=0.10) so we're not just picking up false positives from running 171 tests in parallel. If only the originally-chosen 8 pairs showed dependence, we'd have a cherry-pick problem. If many pairs show dependence, we've discovered a real panel-wide phenomenon.

Headline numbers

Chatterjee-significant
171 / 171
100% pass BH-FDR at q=0.10
Pearson-blind decisive
118 / 171
69% โ€” Spearman |ฯ|<0.30 yet Chatterjee-significant
Production filter
11 / 171
6.4% โ€” what current Pearson baseline flags
Pre-spec threshold
3.5ร—
Observed 69% vs pre-spec 20% threshold

Flagging-bucket comparison

Bucket Count % of 171
Total pairs171100.0%
Spearman |ฯ| โ‰ฅ 0.95 (production "drop")116.4%
Chatterjee ฮพ_max โ‰ฅ 0.10 (empirical decisive)171100.0%
Chatterjee BH-FDR significant at q=0.10171100.0%
Spearman "independent" (|ฯ|<0.30) AND Chatterjee-decisive11869.0%

Top 20 Pearson-blind pairs

These would be retained by the production Pearson filter (low |ฯ|) but show decisive non-linear dependence under Chatterjee + IAAFT. "Spearman says these are unrelated; Chatterjee says one is mostly a curved function of the other."

X Y Spearman ฯ Chatterjee ฮพ_max IAAFT p BH q
kyle_lambda_proxyturnover_imbalanceโˆ’0.1920.7480.00500.0050
kyle_lambda_proxyofiโˆ’0.1920.7480.00500.0050
aggression_ratioprice_impact+0.0760.7330.00500.0050
aggression_ratiobuy_volume+0.1780.7260.00500.0050
aggression_ratioduration_usโˆ’0.1200.7220.00500.0050
lowtrade_intensityโˆ’0.2920.7060.00500.0050
kyle_lambda_proxysell_volume+0.2050.7030.00500.0050
buy_volumeofi+0.2330.6840.00500.0050
buy_volumeturnover_imbalance+0.2330.6820.00500.0050
duration_usofiโˆ’0.1450.6820.00500.0050
duration_usturnover_imbalanceโˆ’0.1450.6800.00500.0050
closetrade_intensityโˆ’0.2860.6760.00500.0050
duration_uskyle_lambda_proxy+0.1750.6730.00500.0050
ofiprice_impact+0.0800.6660.00500.0050
kyle_lambda_proxyprice_impactโˆ’0.1340.6640.00500.0050
price_impactturnover_imbalance+0.0800.6640.00500.0050
aggregation_densityduration_usโˆ’0.2100.6580.00500.0050
aggression_ratiokyle_lambda_proxyโˆ’0.0840.6570.00500.0050
duration_ushigh+0.2160.6430.00500.0050
highvwap_close_deviationโˆ’0.1230.6400.00500.0050

The honest caveats

"100% flagged" โ‰  "drop them all"

Significance is not utility. Many dependencies are statistically significant only because N=1000. Feature selection requires combining dependence + relevance-to-target.

Substrate degeneracy possibility

At 100 dbps, BTC bars are dominated by 1-2 trade events. Many features collapse onto the same trade stream. "100% flagged" could partly reflect substrate physics.

Broadening surface = 171, not 1,378

36 of 53 production features are INSUFFICIENT-CRYPTO-SPARSE at 100 dbps. Realistic surface is the 17-19 reliably-populated features.

Provenance: committed in f4235b13 ยท "multi-symbol robustness + Chatterjee broadening โ€” both validate"