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 ›› 2025, Vol. 33 ›› Issue (8): 37-49.doi: 10.16381/j.cnki.issn1003-207x.2023.0327

• • 上一篇    下一篇

国际新兴资产与中国传统资产的多维溢出效应

王雄, 李景瑶, 任晓航(), 王宗润   

  1. 中南大学商学院,湖南 长沙 410083
  • 收稿日期:2023-03-02 修回日期:2023-08-18 出版日期:2025-08-25 发布日期:2025-09-10
  • 通讯作者: 任晓航 E-mail:domrxh@outlook.com
  • 基金资助:
    国家自然科学基金青年项目(72403258);国家自然科学基金重大项目(72091515)

Multidimensional Spillover of International Emerging Assets and Chinese Traditional Assets: Based on the Quantile VAR Network

Xiong Wang, Jingyao Li, Xiaohang Ren(), Zongrun Wang   

  1. School of Business,Central South University,Changsha 410083,China
  • Received:2023-03-02 Revised:2023-08-18 Online:2025-08-25 Published:2025-09-10
  • Contact: Xiaohang Ren E-mail:domrxh@outlook.com

摘要:

经济全球化背景下的金融市场风险呈现出跨地区溢出和交叉性传染的新特征。本文基于GARCHSK模型和分位数VAR网络框架深入考察了国际新兴资产(金融科技、比特币和绿色债券)与中国传统资产(股市、原油、商品和金属市场)之间在不同市场条件下的多维风险溢出。首先,静态溢出结果表明,资产间的溢出效应是非对称的,极端分位数处的溢出程度普遍高于中分位数,并且不同维度下的左右尾溢出效应也存在明显差异。其次,滚动窗口分析显示,资产价格的收益、波动、偏度和峰度溢出效应均呈现出显著的时变特征,并且这种溢出关系随时间变化的规律更多是受到外部冲击的驱动。最后,通过网络分析方法,发现各金融资产间定向溢出网络的结构特征、传染方向、作用强度和风险中心在不同矩层面和不同分位数下具有异质性。总之,本文的发现有利于进一步厘清国内外金融资产间的多维溢出关系和风险传染路径,对投资者构建多元化资产组合以及决策者加强宏观审慎监管具有深远意义。

关键词: 金融资产, 多维溢出效应, GARCHSK模型, QVAR模型, 非线性网络

Abstract:

Financial market risks in the context of economic globalization have taken on new characteristics of cross-regional spillovers and cross-contagion. An in-depth examination of multidimensional risk spillovers between international emerging assets (fintech, bitcoin and green bonds) and domestic traditional assets (stock market, crude oil, commodity and metal markets) under different market conditions is provided based on the GARCHSK model and the quantile VAR network framework. Firstly, the static results suggest that spillovers between assets are asymmetric, with spillovers generally higher at the extreme quartiles than at the middle quartile, and significant differences in left- and right-tail spillovers across dimensions. Secondly, rolling window analysis shows that asset price returns, volatility, skewness and kurtosis spillovers all exhibit significant time-varying characteristics, and that the pattern of such spillover relationships over time is more driven by external shocks. Finally, through network structure analysis, it is found heterogeneity in the structural characteristics, direction of contagion, intensity of action and risk centres of the targeted spillover network across financial assets at different moments and at different quantiles. Overall, the findings of this paper contribute to further understanding of the multidimensional spillover relationships and risk contagion paths of domestic and foreign financial assets, and have important implications for investors in constructing diversified asset portfolios and policymakers in strengthening macroprudential regulation.

Key words: financial assets, multidimensional spillover effect, GARCHSK model, QVAR model, nonlinear network

中图分类号: