Accuracy — gru_s5
validation only · 2.3M
Summary
| horizon | days | token-days | rank IC | t | LS decile | hit (alpha) | hit (price) | MAE (alpha) | MAE (price) |
|---|---|---|---|---|---|---|---|---|---|
| +1d | 265 | 179k | +0.0727 | 13.0 | +0.82% | 55.9% | 54.4% | 0.0355 | 5.787356 |
| +7d | 265 | 178k | +0.0855 | 17.0 | +2.43% | 61.8% | 61.8% | 0.0933 | 16.535640 |
| +30d | 265 | 177k | +0.1082 | 19.7 | +8.01% | 62.0% | 70.4% | 0.2084 | 33.102553 |
By month
| month | IC +1d | hit +1d | LS +1d | IC +7d | hit +7d | LS +7d | IC +30d | hit +30d | LS +30d | days |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-01 | +0.075 | 56.0% | +0.61% | +0.088 | 63.1% | +0.95% | +0.000 | 47.4% | +3.96% | 28 |
| 2025-12 | +0.049 | 56.2% | +0.94% | +0.048 | 60.3% | +3.28% | +0.036 | 64.1% | +2.63% | 31 |
| 2025-11 | +0.084 | 51.4% | +1.49% | +0.096 | 45.8% | +4.28% | +0.092 | 63.1% | +8.84% | 30 |
| 2025-10 | +0.064 | 59.2% | +0.67% | +0.061 | 66.3% | +1.08% | +0.146 | 45.3% | +17.72% | 31 |
| 2025-09 | +0.069 | 58.9% | +0.44% | +0.076 | 71.3% | +1.84% | +0.105 | 85.1% | +5.92% | 30 |
| 2025-08 | +0.062 | 49.8% | +0.43% | +0.052 | 49.3% | +0.70% | +0.099 | 58.0% | +2.42% | 31 |
| 2025-07 | +0.081 | 53.5% | +0.94% | +0.098 | 53.8% | +2.35% | +0.116 | 51.3% | +2.89% | 31 |
| 2025-06 | +0.081 | 59.6% | +0.83% | +0.097 | 74.1% | +1.70% | +0.142 | 64.2% | +8.20% | 30 |
| 2025-05 | +0.096 | 59.8% | +1.08% | +0.180 | 76.7% | +6.67% | +0.269 | 84.5% | +22.95% | 23 |
t-stats at +7d and +30d are optimistic: consecutive windows overlap, so the effective sample size is closer to the number of distinct days than the number of rows. +1d is the clean read. Long-short spreads are before trading costs and slippage, and the tail of the top-1000 by volume is thin.