Get Started
Research questionWhen building portfolios from predicted returns, should machine-learning models use classification or regression targets?Portfolio models can predict continuous returns or classify securities into return groups. These choices may extract different information and translate differently into investment portfolios.
AI
Finance
Machine Learning
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.Machine Learning Classification and Portfolio Construction: Does the Loss Function Matter?The paper conducts matched empirical comparisons of classification and regression across several model families and examines value weighting, transaction costs, class granularity, subsamples, and spanning tests. Its conclusions are limited to the studied financial data and portfolio-design choices.research paper · Sep 4, 2026
Related questions
How can teams estimate whether multiclass accuracy gains reduce optimization regret before building a model?How can machine-learning performance-modeling frameworks stay maintainable as models and systems evolve?How can machine learning learn from contested labels without assuming one objectively correct target?How should multi-expert models balance diversity and prediction variance for long-tailed visual recognition?
Home
Topics
Search
Library