Mechanism diagram · Tabular Tree
GRADIENT-BOOSTED TREES Tabular features funding · OI · vol · … Tree #1 weak learner Tree #2 weak learner Tree #N weak learner ... boosted residually ... Aggregate sum of tree outputs Lambdarank loss cross-sectional ranks Output scores

How it works

Standard LightGBM with mean-squared-error loss per coin.

The model is trained per-coin (independent) or as a panel; either way the loss is regression, not ranking.

Output is a per-coin forecast.

Pros and cons on this universe

Pros

  • Fast, parallelisable, no GPU needed.
  • Strong on older CPCV consolidated cards (SR_net mean ~+2.2 — but on a different window and different metric).

Cons / failure modes

  • Near-zero universe IC (+0.023).
  • Superseded by LGBM Rank for cross-sectional ranking tasks.
  • Regression loss does not align with the cross-sectional decision the strategy makes.

References