Portfolio Optimization for Pension Purposes: Literature Review
系统综述了82篇近十年文献，指出养老金投资组合优化模型在实施中面临缺乏针对性模型、忽视长寿风险和可持续性约束等挑战，呼吁引入机器学习、ESG框架以提升模型实用性。
ABSTRACT This systematic review identifies persistent challenges and gaps in the literature on pension portfolio optimization models. We searched, selected, and critically analyzed 82 articles from three major academic databases published over the past decade to investigate the barriers to the effective implementation of these models. Most studies primarily focus on traditional portfolio optimization approaches, often overlooking recent innovations and the unique structural characteristics of pension systems. While the number of publications has grown steadily over the past 10 years, reflecting increasing academic interest in the topic, our analysis reveals a continued need for substantive research advances to bridge existing gaps. Key challenges include the lack of models tailored to the specific features of pension funds, limited attention to longevity and sustainability constraints, and insufficient application of innovative methods to address sector‐specific demands. This study highlights the importance of future research that incorporates advanced machine learning techniques, develops ESG‐aligned and sustainability‐oriented optimization frameworks, and more effectively models regulatory and demographic constraints. These directions are essential for enhancing the robustness and relevance of pension portfolio models and to support more effective, long‐term pension fund management.