Sticker | Lucas (Foil)
7-day forecast validation for the Sticker category, 21 items (walk-forward, 47 folds)
| Model | Mean abs. error | vs. no change | Folds won | Range coverage (calibrated) | Interval score | Direction accuracy |
|---|---|---|---|---|---|---|
| No change (baseline) | 0.0854 | baseline | — | 80.8% | 0.4766 | — |
| Average drift | 0.0856 | -0.3% | 20/47 | 80.8% | 0.4952 | 49.9% |
| AR(1) momentum/reversion | 0.0849 | +0.6% | 33/47 | 80.6% | 0.4923 | 53.3% |
No model beats “no change” significantly and consistently (best: AR(1) momentum/reversion, t=0.99, 33/47 folds won). Errors are on log-price changes; significance is tested on per-date averages with Newey-West standard errors. Range widths are calibrated on earlier folds only; the closer coverage is to 80%, the more trustworthy the range. The interval score rewards ranges that are both narrow and right (lower is better).
30-day forecast validation for the Sticker category, 21 items (walk-forward, 47 folds)
| Model | Mean abs. error | vs. no change | Folds won | Range coverage (calibrated) | Interval score | Direction accuracy |
|---|---|---|---|---|---|---|
| No change (baseline) | 0.1537 | baseline | — | 80.5% | 0.9431 | — |
| Average drift | 0.1559 | -1.5% | 16/47 | 79.3% | 0.9158 | 49.7% |
| AR(1) momentum/reversion | 0.1561 | -1.6% | 17/47 | 79.4% | 0.9152 | 49.9% |
No model beats “no change” significantly and consistently (best: Average drift, t=1.89, 16/47 folds won). Errors are on log-price changes; significance is tested on per-date averages with Newey-West standard errors. Range widths are calibrated on earlier folds only; the closer coverage is to 80%, the more trustworthy the range. The interval score rewards ranges that are both narrow and right (lower is better).