Accuracy — gru_s8
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 | +0.0769 | 13.0 | +0.83% | 55.9% | 54.4% | 0.0355 | 5.767054 |
| +7d | 265 | 178k | +0.0928 | 16.2 | +2.46% | 60.9% | 61.2% | 0.0934 | 16.223175 |
| +30d | 265 | 177k | +0.1185 | 22.1 | +8.46% | 61.1% | 69.5% | 0.2085 | 32.529890 |
By month
| month | IC +1d | hit +1d | LS +1d | IC +7d | hit +7d | LS +7d | IC +30d | hit +30d | LS +30d | days |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-01 | +0.080 | 56.1% | +0.86% | +0.091 | 63.3% | +2.22% | -0.014 | 47.7% | +6.16% | 28 |
| 2025-12 | +0.055 | 56.4% | +0.96% | +0.078 | 60.0% | +5.10% | +0.081 | 63.6% | +9.69% | 31 |
| 2025-11 | +0.081 | 51.3% | +1.62% | +0.097 | 44.2% | +5.52% | +0.125 | 62.1% | +17.20% | 30 |
| 2025-10 | +0.065 | 58.9% | +0.74% | +0.058 | 65.8% | +0.11% | +0.150 | 44.8% | +15.11% | 31 |
| 2025-09 | +0.067 | 58.9% | +0.25% | +0.047 | 71.3% | -0.39% | +0.072 | 85.0% | -3.27% | 30 |
| 2025-08 | +0.073 | 49.8% | +0.46% | +0.085 | 49.3% | +1.05% | +0.126 | 58.0% | +4.26% | 31 |
| 2025-07 | +0.091 | 53.6% | +1.03% | +0.108 | 53.8% | +2.86% | +0.113 | 51.2% | +4.42% | 31 |
| 2025-06 | +0.097 | 59.8% | +0.94% | +0.125 | 73.5% | +1.70% | +0.169 | 63.8% | +10.33% | 30 |
| 2025-05 | +0.088 | 59.2% | +0.58% | +0.166 | 69.7% | +4.47% | +0.271 | 76.9% | +13.17% | 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.