Paper: arXiv 2610.26456

Authors: Kelvin J. L. Koa, Ziling Liu, Xinyang Li, Shuowei Feng, Ruoyu Liang, Mathieu Giroux, Eugene Y. Wen, Xuan Yao, Ke-Wei Huang, Hugo Lamarre

Abstract

Sector rotation is a common decision in benchmark-relative portfolio management, where portfolio managers rank sectors for overweight and underweight positions. We formulate this task as ranking sector actions by expected future risk-adjusted active performance, measured with forward sector-benchmark Sharpe ratios. Such rankings should preserve temporal information, express model outputs through economically familiar signal characteristics, account for changes in signal relevance across macroeconomic environments, and capture dependencies among sectors. We propose RegimeRank, an interpretable sector-ranking model that represents fundamental and technical signals through expert-defined temporal characteristics, conditions them on learned macro factors via cross-attention, and models cross-sector dependencies through self-attention. Expected-gradient attributions identify the contributions of individual signal-transformation pairs, while macroeconomic and cross-sector attention maps provide diagnostics of the learned conditioning and relational structure. We evaluate RegimeRank on weekly S&P 500 data from 2002 to 2024 under combinatorial purged cross-validation, where it outperforms representative ranking and machine-learning baselines across multiple ranking metrics. Ablation studies support the contributions of latent macro states, regime conditioning, and cross-sector modelling, while attribution and attention analyses illustrate the temporal characteristics, macro states, and sector relationships underlying its rankings.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a sophisticated model combining multiple advanced AI techniques for a complex financial problem. It demonstrates strong empirical validation with extensive testing and ablation studies, suggesting a robust and novel approach to sector rotation.

Research Flowchart

  flowchart TD
    A[Research Goal: Interpretably Rank Sectors for Rotation] --> B{Key Methodology: RegimeRank Model};

    B --> C[Data/Inputs: Weekly S&P 500 (2002-2024), Fundamental & Technical Signals];

    C --> D{Computational Process: RegimeRank Model};
    D --> D1[Expert-defined Temporal Characteristics];
    D --> D2[Cross-Attention for Macro Conditioning];
    D --> D3[Self-Attention for Cross-Sector Dependencies];
    D --> D4[Expected-Gradient Attributions & Attention Maps];

    D --> E[Key Findings/Outcomes];
    E --> F["Outperforms Baselines (Ranking & ML)"];
    E --> G["Attribution & Attention Analyses Provide Interpretability"];
    E --> H["Ablation Studies Validate Latent Macro States, Regime Conditioning, Cross-Sector Modeling"];