Research map / Learning theory and algorithms / Kernels and regression theory
Loss Control with Rank-one Covariance Estimate for Short-term Portfolio Optimization
Zhao-Rong Lai, Liming Tan, Xiaotian Wu, Liangda Fang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the rank-one covariance risk term (gamma > 0) (ablation) | SPOLC · FOF | cumulative wealth: 1.9228 vs gamma = 0 worse | Improved |
| use SPOLC loss control instead of original mean-variance model | portfolio optimization · FOF | cumulative wealth: 1.3881 -> 1.9228 | Improved |
| use rank-one covariance estimate instead of Gram matrix S (ablation) | SPOLC · SP500 | cumulative wealth: 43.7565 -> 58.1662 | Improved |
| use rank-one covariance estimate instead of Gram matrix S (ablation) | SPOLC · NYSE19 | cumulative wealth: 3.4868 -> 2.4530 | Hurt |
| use rank-one covariance estimate instead of NSCM (ablation) | SPOLC · TSE | cumulative wealth: 1.5738 -> 345.1481 | Improved |
| use rank-one covariance estimate instead of NSCM (ablation) | SPOLC · FRENCH | cumulative wealth: 6.5739 vs 20.4540 | Hurt |
| increase NSCM window size from 5 to 300 (ablation) | NSCM · FOF | cumulative wealth: 1.1134 -> 1.1349 | No clear effect |
| increase window size w from 5 to 10 (ablation) | SPOLC · FOF | cumulative wealth: 1.9228 -> 0.5971 | Hurt |
| use SPOLC instead of competing SPO systems for VaR | SPOLC · FOF VaR c=90% | value at risk: -0.0243 -> -0.0166 | Improved |
| use SPOLC instead of competitors for maximum drawdown | SPOLC · NYSE19 | maximum drawdown: 0.8357 -> 0.7186 | Improved |
| use SPOLC instead of AICTR for maximum drawdown | SPOLC · DJIA | maximum drawdown: 0.3324 -> 0.3552 | Hurt |
Closest papers by meaning
- Latent Variable Estimation in Bayesian Black-Litterman Models (ICML 2025 · similarity 0.83)
- Retrieval-Augmented Diffusion Modeling for Stochastic Discount Factor Portfolios (NeurIPS 2026 · similarity 0.83)
- A Set-Sequence Model for Time Series (NeurIPS 2026 · similarity 0.82)
- Decision Making under the Exponential Family: Distributionally Robust Optimisation with Bayesian Ambiguity Sets (ICML 2025 · similarity 0.82)
- On the Tightness and Computational Tractability of Higher-Dimensional Confidence Sequences (NeurIPS 2026 · similarity 0.82)
- Towards Assessing and Benchmarking Risk-Return Tradeoff of Off-Policy Evaluation (ICLR 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
