Reproduce “Decision-focused Sparse Tangent Portfolio Optimization” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Haeun Jeon, Seunghoon Choi, Hyunglip Bae, Yongjae Lee, Woo Chang Kim
The authors train a return predictor through a differentiable portfolio pipeline: it forecasts returns, applies a soft top-k mask with total mass k, and reoptimizes weights. On KOSPI200, 162 assets with a 16-asset budget produced mean daily out-of-sample Sharpe 1.958 ± 0.022 versus 1.372 ± 0.052 for mSSRM-PGA over five seeds; SD-relaxation PFL led in EuroStoxx50 and some FTSE100 settings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.