Rolling projection DOE
Find the projection setup whose median is closest to actual outcomes.
Replay completed anchors and rank horizon × sample-size settings by projection error.
Best rolling setting
3D · n=80
DOE results snapshot · Window: 2026-01-05 → 2026-01-26 · anchors: 60 · latest snapshot: 2026-01-29 · computed at: 2026年7月9日 下午7:01 HKT
DOE setting comparison
Same idea as score-vs-outcome analysis: compare each parameter setting by observed actual-minus-median error over rolling completed anchors. Lower median abs error and lower bias win.
Best setting 3D · n=80: median projection vs actual return
X = Median projection return. Y = actual return. Solid line = linear trend. Orange line = bucket average. Checked: 48 · Pending: 0 · Avg: -0.36%
Best setting 3D · n=80: median projection vs forecast error
X = Median projection return. Y = actual - median. Solid line = linear trend. Orange line = bucket average. Checked: 48 · Pending: 0 · Avg: -0.73%
DOE ranking
Ranked by a composite DOE score: lower median/mean absolute error, lower bias, enough anchors, and better IQR coverage.
| Rank | Horizon | n | Median abs | Mean abs | Bias | Inside P25/P75 | Anchors | DOE score |
|---|---|---|---|---|---|---|---|---|
| #1 | 3D | 80 | 1.55% | 2.29% | -0.73% | 50% | 48 | 1.561 |
| #2 | 3D | 160 | 1.61% | 2.24% | -0.73% | 50% | 48 | 1.574 |
| #3 | 3D | 120 | 1.68% | 2.28% | -0.76% | 52.1% | 48 | 1.622 |
| #4 | 3D | 240 | 1.65% | 2.28% | -0.81% | 47.9% | 48 | 1.625 |
| #5 | 3D | 500 | 1.65% | 2.33% | -0.86% | 52.1% | 48 | 1.639 |
| #6 | 3D | 320 | 1.71% | 2.3% | -0.82% | 47.9% | 48 | 1.663 |
| #7 | 3D | 40 | 1.83% | 2.3% | -0.65% | 47.9% | 48 | 1.698 |