欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2026, Vol. 43 ›› Issue (6): 728-736.DOI: 10.7523/j.ucas.2025.002

• 数学与物理学 • 上一篇    下一篇

HN-GARCH模型下时间一致的均值-方差投资组合

阮中杰1, 罗翠翠2()   

  1. 1.中国科学院大学数学科学学院,北京 100049
    2.中国科学院大学国际学院,北京 100190
  • 收稿日期:2024-10-15 修回日期:2025-10-01 发布日期:2026-11-19
  • 通讯作者: 罗翠翠
  • 基金资助:
    国家自然科学基金(72210107001);北京市自然科学基金(IS23128);中央高校基本科研业务费和中国科学院PIFI国际杰出团队项目(2024PG0013)

Time-consistent mean-variance portfolio selection under HN-GARCH

Zhongjie RUAN1, Cuicui LUO2()   

  1. 1.School of Mathematical Sciences,University of Chinese Academy of Sciences,Beijing 100049,China
    2.International College,University of Chinese Academy of Sciences,Beijing 100190,China
  • Received:2024-10-15 Revised:2025-10-01 Published:2026-11-19
  • Contact: Cuicui LUO

摘要:

首先在HN-GARCH模型下推导出均值-方差投资组合问题时间一致的均衡解。然后,以指数效用下的最优解作为参照,引入一个评价资产组合表现的指标,即确定性等价回报损失。最后,基于纳斯达克100股指的历史数据,蒙特卡洛模拟表明HN-GARCH模型下的时间一致的资产组合在CErL和夏普率的表现上优于同方差模型下的表现,非对称型HN-GARCH模型的表现优于对称型HN-GARCH。

关键词: HN-GARCH模型, 均值-方差, 时间一致, 指数效用, 确定性等价回报损失

Abstract:

In this paper, we derive a time-consistent solution for mean-variance portfolio selection within the framework of the HN-GARCH model. Then, by using the optimal solution derived under exponential utility as a reference point, we introduce a well-defined performance metric, namely certainty equivalent return loss (CErL), to assess portfolio performance. Finally, based on the historical data of the NASDAQ-100 Index, Monte Carlo simulations empirically demonstrate that the solution under the HN-GARCH model outperforms that under a homoscedastic variant in terms of CErL and the performance of asymmetric HN-GARCH is better than symmetric HN-GARCH’s performance.

Key words: HN-GARCH model, mean-variance, time-consistent, exponential utility, certainty equivalent return loss

中图分类号: