主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2026, Vol. 34 ›› Issue (7): 33-48.doi: 10.16381/j.cnki.issn1003-207x.2023.1100

• • 上一篇    下一篇

多元GARCH-Ito^模型及其在高维波动率矩阵预测中的应用

宋馨雨1, 邓媛元1, 周勇2, 苑慧玲3()   

  1. 1.上海财经大学统计与数据科学学院,上海 200433
    2.华东师范大学统计学院,上海 200062
    3.东华大学数学与统计学院,上海 201620
  • 收稿日期:2023-06-29 修回日期:2024-09-14 出版日期:2026-07-25 发布日期:2026-06-18
  • 通讯作者: 苑慧玲 E-mail:yuanhuiling2016@gmail.com
  • 基金资助:
    国家自然科学基金项目(72103118);国家自然科学基金项目(72403089);国家自然科学基金项目(72531003);中国博士后科学基金项目(2023M741190);上海市白玉兰人才计划浦江项目(23PJ1402400);2025年上海市白玉兰人才计划青年项目;上海浦江人才计划(21PJC056)

Multivariate GARCH-Ito^Model with Applications in High-dimension Volatility Matrix Prediction

Xinyu Song1, Yuanyuan Deng1, Yong Zhou2, Huiling Yuan3()   

  1. 1.School of Statistics and Data Science,Shanghai University of Finance and Economics,Shanghai 200433,China
    2.School of Statistics,East China Normal University,Shanghai 200062,China
    3.School of Mathematics and Statistics,Donghua University,Shanghai 201620,China
  • Received:2023-06-29 Revised:2024-09-14 Online:2026-07-25 Published:2026-06-18
  • Contact: Huiling Yuan E-mail:yuanhuiling2016@gmail.com

摘要:

本文结合高频和低频数据,提出多元GARCH-Ito^模型,该模型将多元GARCH结构嵌入到连续伊藤扩散过程的瞬时波动率中,能够较好地刻画资产波动率矩阵的动态特征。本文利用拟极大似然方法对模型参数进行估计,给出了统计量的相关渐近性质。当涉及高维资产时,将因子模型与多元GARCH-Ito^模型相结合,提出了高维积分波动率矩阵的预测方法,并给出了预测统计量的统计性质。并且,分别在低维和高维的情形下,利用模拟分析对模型参数估计和波动率矩阵预测方法的统计性质进行了验证。最后,选择2018年1月1日—2020年12月31日沪深300指数的270只成分股进行实证分析,与现有的波动率模型比较,本文提出的多元GARCH-Ito^模型在积分波动率矩阵预测和投资组合分配中都具有优良的表现,从而表明本文利用高频和低频数据构建的波动率模型具有实际的应用价值。

关键词: 高频数据, 动态结构, 拟极大似然方法, 积分波动率矩阵, 因子模型

Abstract:

A novel multivariate GARCH-Ito^ model is introduced that integrates the structural framework of multivariate GARCH within a continuous-time diffusion process, providing a unified approach to modeling the dynamic evolution of volatility matrices by jointly utilizing high-frequency and low-frequency data. A quasi-maximum likelihood estimation method for parameter inference is developed and the corresponding asymptotic theory is established. To address the challenges associated with high-dimensional asset spaces, the proposed model is coupled with a factor structure, enabling scalable and efficient prediction of large integrated volatility matrices. Theoretical guarantees for the proposed predictor are provided under high-dimensional settings. Extensive simulation studies are conducted to examine the finite-sample performance of both the estimation and prediction procedures across both low-dimensional and high-dimensional contexts. In an empirical application, 270 constituent stocks of the CSI 300 index are analyzed using minute-level high-frequency data from January 1, 2018, to December 31, 2020. The results demonstrate that the multivariate GARCH-Ito^ model consistently outperforms several benchmark methods in terms of integrated volatility matrix forecasting and portfolio allocation, offering a flexible and unified framework that incorporates both high- and low-frequency data features for advanced volatility modeling and prediction.

Key words: high-frequency data, dynamic structure, quisi-maximum likelihood method, integrated volatility matrix, factor model

中图分类号: