Long/flat, held [t,t+1], 2bps/flip. Per-leg block placebo (21d, 800) + Bonferroni over the family + DSR. CANDIDATE bars: OOS net > B&H, p < Bonferroni, IS edge same sign, partial-IC|mom > 0.
SPY (26 trials, Bonferroni bar p < 0.0019)
| signal | OOS net SR | OOS B&H | IS net | IS B&H | partial-IC | placebo p | flag |
|---|---|---|---|---|---|---|---|
| vix_lvl:low | 0.94 | 0.80 | 0.19 | 0.33 | +0.007 | 0.0587 | - |
| vix_rank:low | 0.94 | 0.80 | 0.22 | 0.33 | +0.002 | 0.0475 | - |
| ts3m:high | 0.93 | 0.80 | 0.13 | 0.33 | -0.011 | 0.0387 | - |
| ts9d_3m:low | 0.92 | 0.80 | 0.13 | 0.33 | +0.015 | 0.0400 | - |
| tent:low | 0.89 | 0.80 | 0.01 | 0.33 | +0.016 | 0.0488 | - |
| vix_chg5:low | 0.86 | 0.80 | -0.30 | 0.33 | +0.027 | 0.0350 | - |
| vrp:high | 0.76 | 0.80 | 0.68 | 0.33 | +0.029 | 0.1537 | - |
| rv_vix:low | 0.76 | 0.80 | 0.68 | 0.33 | -0.029 | 0.1537 | - |
| vvix_lvl:low | 0.72 | 0.80 | 0.13 | 0.33 | +0.002 | 0.1850 | - |
| vvix_vix:high | 0.69 | 0.80 | 0.39 | 0.33 | -0.013 | 0.3100 | - |
| ts9d:low | 0.60 | 0.80 | 0.02 | 0.33 | +0.012 | 0.2812 | - |
| vix_chg1:high | 0.59 | 0.80 | 0.22 | 0.33 | +0.043 | 0.1650 | - |
| vxx_svxy:low | 0.56 | 0.80 | 0.00 | 0.33 | +0.047 | 0.3625 | - |
| vvix_vix:low | 0.51 | 0.80 | 0.05 | 0.33 | -0.013 | 0.4875 | - |
| ts9d:high | 0.48 | 0.80 | 0.48 | 0.33 | +0.012 | 0.3975 | - |
| vvix_lvl:high | 0.41 | 0.80 | 0.17 | 0.33 | +0.002 | 0.6763 | - |
| vrp:low | 0.39 | 0.80 | -0.23 | 0.33 | +0.029 | 0.7188 | - |
| rv_vix:high | 0.39 | 0.80 | -0.23 | 0.33 | -0.029 | 0.7188 | - |
| vix_lvl:high | 0.37 | 0.80 | 0.16 | 0.33 | +0.007 | 0.6987 | - |
| vix_rank:high | 0.36 | 0.80 | 0.13 | 0.33 | +0.002 | 0.7188 | - |
| ts3m:low | 0.35 | 0.80 | 0.52 | 0.33 | -0.011 | 0.6375 | - |
| vxx_svxy:high | 0.34 | 0.80 | 0.00 | 0.33 | +0.047 | 0.4138 | - |
| ts9d_3m:high | 0.33 | 0.80 | 0.42 | 0.33 | +0.015 | 0.6125 | - |
| tent:high | 0.20 | 0.80 | 0.57 | 0.33 | +0.016 | 0.8113 | - |
| vix_chg5:high | 0.18 | 0.80 | 0.44 | 0.33 | +0.027 | 0.8888 | - |
| vix_chg1:low | 0.14 | 0.80 | -0.22 | 0.33 | +0.043 | 0.8313 | - |
Candidates: 0/26
QQQ (26 trials, Bonferroni bar p < 0.0019)
| signal | OOS net SR | OOS B&H | IS net | IS B&H | partial-IC | placebo p | flag |
|---|---|---|---|---|---|---|---|
| vix_rank:low | 1.55 | 1.36 | 0.09 | 0.25 | +0.000 | 0.0862 | - |
| vrp:high | 1.50 | 1.36 | 0.73 | 0.25 | +0.061 | 0.1100 | - |
| rv_vix:low | 1.50 | 1.36 | 0.73 | 0.25 | -0.061 | 0.1100 | - |
| vix_chg1:high | 1.47 | 1.36 | -0.29 | 0.25 | +0.028 | 0.0387 | - |
| vxx_svxy:low | 1.31 | 1.36 | -0.16 | 0.25 | +0.055 | 0.2725 | - |
| vvix_lvl:low | 1.26 | 1.36 | -0.29 | 0.25 | +0.046 | 0.2475 | - |
| ts9d:high | 1.26 | 1.36 | 0.34 | 0.25 | +0.026 | 0.1625 | - |
| vix_lvl:low | 1.15 | 1.36 | 0.21 | 0.25 | +0.031 | 0.3713 | - |
| vix_chg5:low | 1.10 | 1.36 | -0.32 | 0.25 | +0.027 | 0.2487 | - |
| tent:high | 1.03 | 1.36 | -0.53 | 0.25 | +0.006 | 0.2587 | - |
| ts3m:low | 1.03 | 1.36 | 0.10 | 0.25 | -0.030 | 0.2988 | - |
| vvix_vix:low | 1.01 | 1.36 | -0.20 | 0.25 | -0.003 | 0.3538 | - |
| ts9d_3m:high | 0.95 | 1.36 | 0.48 | 0.25 | +0.035 | 0.3588 | - |
| ts9d_3m:low | 0.90 | 1.36 | -0.54 | 0.25 | +0.035 | 0.5887 | - |
| vvix_vix:high | 0.84 | 1.36 | 0.41 | 0.25 | -0.003 | 0.6475 | - |
| ts3m:high | 0.80 | 1.36 | -0.06 | 0.25 | -0.030 | 0.7137 | - |
| vix_lvl:high | 0.77 | 1.36 | -0.11 | 0.25 | +0.031 | 0.4875 | - |
| tent:low | 0.76 | 1.36 | 0.71 | 0.25 | +0.006 | 0.7025 | - |
| vix_chg5:high | 0.73 | 1.36 | 0.23 | 0.25 | +0.027 | 0.6175 | - |
| vvix_lvl:high | 0.70 | 1.36 | 0.48 | 0.25 | +0.046 | 0.5725 | - |
| vxx_svxy:high | 0.52 | 1.36 | 0.33 | 0.25 | +0.055 | 0.5463 | - |
| vix_rank:high | 0.47 | 1.36 | -0.04 | 0.25 | +0.000 | 0.7688 | - |
| ts9d:low | 0.43 | 1.36 | -0.39 | 0.25 | +0.026 | 0.9000 | - |
| vrp:low | 0.17 | 1.36 | -0.53 | 0.25 | +0.061 | 0.9600 | - |
| rv_vix:high | 0.17 | 1.36 | -0.53 | 0.25 | -0.061 | 0.9600 | - |
| vix_chg1:low | 0.03 | 1.36 | 0.09 | 0.25 | +0.028 | 0.9738 | - |
Candidates: 0/26
Reading
- With this many trials only a p below the Bonferroni bar counts; a raw p<0.05 leg is expected by chance.
- A CANDIDATE is not adopted: it must be re-run frozen (single hypothesis, placebo, IS/OOS, costs) and then shadowed.