Accuracy — lstm_s9
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 | +0.0721 | 13.7 | +0.72% | 55.3% | 54.4% | 0.0356 | 5.713215 |
| +7d | 265 | 178k | +0.0911 | 18.9 | +2.63% | 60.4% | 60.7% | 0.0935 | 16.096402 |
| +30d | 265 | 177k | +0.1062 | 22.8 | +7.17% | 60.8% | 69.1% | 0.2090 | 32.973822 |
By month
| month | IC +1d | hit +1d | LS +1d | IC +7d | hit +7d | LS +7d | IC +30d | hit +30d | LS +30d | days |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-01 | +0.082 | 56.2% | +0.70% | +0.103 | 63.3% | +1.73% | -0.003 | 47.6% | -0.07% | 28 |
| 2025-12 | +0.049 | 55.9% | +0.65% | +0.081 | 60.4% | +4.56% | +0.070 | 64.3% | +5.99% | 31 |
| 2025-11 | +0.076 | 52.0% | +1.39% | +0.082 | 45.8% | +5.21% | +0.112 | 63.4% | +10.46% | 30 |
| 2025-10 | +0.051 | 58.0% | +0.54% | +0.052 | 64.8% | +0.72% | +0.123 | 44.5% | +11.51% | 31 |
| 2025-09 | +0.067 | 59.0% | +0.10% | +0.057 | 71.3% | +0.31% | +0.052 | 85.0% | -1.97% | 30 |
| 2025-08 | +0.093 | 49.9% | +0.65% | +0.104 | 49.4% | +1.69% | +0.146 | 58.0% | +7.00% | 31 |
| 2025-07 | +0.091 | 53.9% | +0.97% | +0.124 | 55.0% | +3.80% | +0.111 | 51.7% | +8.49% | 31 |
| 2025-06 | +0.062 | 59.9% | +0.63% | +0.083 | 74.5% | +1.51% | +0.120 | 64.0% | +6.86% | 30 |
| 2025-05 | +0.083 | 52.6% | +0.83% | +0.148 | 59.1% | +4.53% | +0.252 | 70.9% | +18.22% | 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.