用动态图网络捕捉危机中资产关系变化，提升投资组合抗风险能力。
Crisis-Resilient Portfolio Management via Graph-based Spatio-Temporal Learning
- 通过自注意力机制动态学习资产间关联，自动过滤92.5%噪声连接
- 在通胀危机中仍保持高收益，夏普比率达3.76，优于基线707%
- 可解释的注意力权重揭示危机期防御性资产聚集特征
金融时间序列预测面临核心挑战：最优资产配置需理解随危机演变的制度依赖型相关结构。现有基于图的时空学习方法依赖预设图拓扑（如相关阈值、行业分类），在信用传染、疫情冲击或通胀抛售等不同危机机制下难以适应。本文提出CRISP（Crisis-Resilient Investment through Spatio-temporal Patterns）框架，结合图卷积网络建模空间关系、双向LSTM加自注意力捕捉时序动态，并通过多头图注意力网络学习稀疏结构。相比固定拓扑方法，CRISP通过注意力机制自主识别关键资产关系，过滤92.5%连接作为噪声，同时保留危机相关依赖以实现精准的制度特异性预测。该模型基于2005–2021年数据训练，涵盖信用与疫情危机，在2022–2024年通胀驱动市场（根本不同制度）中表现出强泛化能力，准确预测相应相关结构。由此实现动态投资组合配置，维持下行周期盈利能力，夏普比率达3.76，较均权基准提升707%，较静态图方法提升94%。学习到的注意力权重提供可解释的制度识别：危机期间防御性集群注意力强度提升49%，远高于全市场平均31%，为学习预测而非预设假设带来的涌现行为。
原文摘要 · Abstract (English)
Financial time series forecasting faces a fundamental challenge: predicting optimal asset allocations requires understanding regime-dependent correlation structures that transform during crisis periods. Existing graph-based spatio-temporal learning approaches rely on predetermined graph topologies--correlation thresholds, sector classifications--that fail to adapt when market dynamics shift across different crisis mechanisms: credit contagion, pandemic shocks, or inflation-driven selloffs. We present CRISP (Crisis-Resilient Investment through Spatio-temporal Patterns), a graph-based spatio-temporal learning framework that encodes spatial relationships via Graph Convolutional Networks and temporal dynamics via BiLSTM with self-attention, then learns sparse structures through multi-head Graph Attention Networks. Unlike fixed-topology methods, CRISP discovers which asset relationships matter through attention mechanisms, filtering 92.5% of connections as noise while preserving crisis-relevant dependencies for accurate regime-specific predictions. Trained on 2005--2021 data encompassing credit and pandemic crises, CRISP demonstrates robust generalization to 2022--2024 inflation-driven markets--a fundamentally different regime--by accurately forecasting regime-appropriate correlation structures. This enables adaptive portfolio allocation that maintains profitability during downturns, achieving Sharpe ratio 3.76: 707% improvement over equal-weight baselines and 94% improvement over static graph methods. Learned attention weights provide interpretable regime detection, with defensive cluster attention strengthening 49% during crises versus 31% market-wide--emergent behavior from learning to forecast rather than imposing assumptions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。