Accurate transfer of information across multiple sectors to enhance model estimation is both significant and challenging in multi-sector portfolio optimization involving a large number of assets in different classes. Within the framework of factor modeling, we propose a novel data-adaptive multi-task learning methodology that quantifies and learns the relatedness among the principal temporal subspaces (spanned by factors) across multiple sectors under study. This approach not only improves the simultaneous estimation of multiple factor models but also enhances multi-sector portfolio optimization, which heavily depends on the accurate recovery of these factor models. Additionally, a novel and easy-to-implement algorithm, termed projection-penalized principal component analysis, is developed to accomplish the multi-task learning procedure. Diverse simulation designs and practical application on daily return data from Russell 3000 index demonstrate the advantages of multi-task learning methodology.

翻译：跨领域准确传递信息以增强模型估计，对于涉及大量不同类别资产的多领域投资组合优化而言，既至关重要又充满挑战。在因子模型框架下，我们提出了一种新颖的数据自适应多任务学习方法，该方法能够量化并学习所研究的多个领域中主要时间子空间（由因子张成）之间的关联性。这一方法不仅改进了多个因子模型的同步估计，还提升了高度依赖这些因子模型精准恢复的多领域投资组合优化效果。此外，我们开发了一种新颖且易于实现的算法——投影惩罚主成分分析，以完成多任务学习过程。基于Russell 3000指数日收益率数据的多样化模拟设计与实际应用，共同验证了多任务学习方法的优势。