Research brief 006

What if the model learned the portfolio instead of the forecast?

KellyBoost skips the forecast-then-optimize sequence and trains directly on wealth growth. The objective changes the bet.

The idea worth investigating

A good forecast and a good allocation are different objectives. KellyBoost brings the allocation decision into model training, using boosted trees to choose weights rather than merely forecast a label.

The full analysis examines the study and the practical question behind it: KellyBoost skips the forecast-then-optimize sequence and trains directly on wealth growth. The objective changes the bet.

Paper-based analysis; no independent house test has been completed.

Paid analysis on Substack

See what changes the investment case.

The headline is the starting point. The subscriber analysis takes you through:

  • The direct-growth versus surrogate comparison
  • Concentration and estimation risk
  • The benchmark and cost evidence that determine our assessment
Read the paid analysis ↗

AI assists curation and drafting. The research status distinguishes paper reviews from house tests. Read our research approach.

New research briefs by email, free

Find your next research question.

Get the finding that caught our attention, the original paper, and the practical question it raises. Full reviews and house-test details are available with a paid subscription.