基于风险传导网络的投资组合自抗扰研究

doi:10.16381/j.cnki.issn1003-207x.2022.2439

摘要/Abstract

摘要：

均值方差（MV）模型是现代资产配置理论的基础，通过最小化投资组合内资产收益相关性以分散风险。但是，MV受到建模随机误差和市场内外部风险的干扰，模型泛化性能较差。现有投资组合研究主要通过修正风险度量、优化风险结构的方式改进MV建模误差，但模型仍具有强数据依赖，局限于风险分散，无法抵御外生冲击。风险传导具有局部确定性因果关系，但经过多重路径传导形成了复杂关联，使风险总体行为呈现高阶非线性特性。本文提出风险自抗扰的改进思路，利用风险相继作用的确定性，改进MV得到投资组合风险自抗扰模型（portfolio risk active disturbance rejection model，PRADR）。基于A股市场的实证结果显示，股市风险由独立风险根因高阶交互沿供应链相继传导形成，高阶风险传导网络仅需一阶线性网络风险根因数目的 $1 / 7$ ，即达到相同水平的风险解释度；经过二维交叉重复试验，PRADR夏普比率更大且波动更小，投资组合表现优于MV；面对股市外生冲击，PRADR风险自抗扰能力强，具有更高的泛化性能。

关键词: 均值方差模型, 风险自抗扰, 风险传导网络, 独立成分分析

Abstract:

The mean variance （MV） model is the basis of modern asset allocation theory. MV disperses risk through minimizing the correlation of asset returns within the portfolio. Since the variance is not applicable to the non-Gaussian distribution of asset returns and the covariance can only describe the linear correlation between assets， MV is interfered by modeling random errors and the internal or external risks in the market， resulting in poor generalization performance. The existing portfolio researches mainly improve the MV modeling error by modifying the risk measurement and optimizing the risk structure， but those models still have strong data dependence which makes the risk offset unstable. Besides， those models are limited to risk dispersion which cannot resist exogenous shocks. Risk conduction exists partially deterministic causal relationship. However， financial risk forms a complex association through multiple paths of conduction， which makes the overall behavior show high order nonlinear characteristics. In this paper， an improved idea of risk auto disturbance rejection is proposed. Using the sequential risk actions， MV is improved to construct Portfolio Risk Active Disturbance Rejection Model （PRADR）. Empirical research is conducted using A-share stock market data from July 10， 2017 to December 30， 2022， and the results show that the stock market risk is formed by the high-order of independent risk causes interacted and sequentially conducted along the supply chain. The high-order risk conduction network only needs $1 / 7$ risk causes of the first-order linear network， which achieves the same risk interpretation degree； Through two-dimensional cross repeated test， PRADR Sharpe ratio is larger and less volatile， which illustrates its portfolio performance is better than MV； Suffering from the exogenous stock， PRADR has stronger risk auto disturbance rejection ability and higher generalization performance. The key to portfolio risk autoimmunity is to extract the deterministic relationships contained in uncertain risks and resist uncertainty interference. The deterministic risk transmission laws contained in various enterprise relationships need to be explored and applied.

Key words: mean-variance model, risk active disturbance rejection, risk conduction network, independent component correlation algorithm

中图分类号:

F832

陈苑榕, 宋海涛. 基于风险传导网络的投资组合自抗扰研究[J]. 中国管理科学, 2025, 33(4): 50-61.

Yuanrong Chen, Haitao Song. Research on Portfolio Auto Disturbance Rejection Based on Risk Conduction Network[J]. Chinese Journal of Management Science, 2025, 33(4): 50-61.

图/表 13

图1

图2

图3

图4

图5

图6

图7

表1

一阶高阶风险贡献率（由左至右1阶贡献率递减）"

风险延伸因子	$f 5$	$f 7$	$f 1$	$f 2$	$f 4$	$f 9$	$f 10$	$f 8$	$f 11$	$f 3$	$f 6$
一阶贡献率（%）	64.9	57.2	50.7	43.1	39.7	26.6	23.3	14.7	7.6	7.4	5.7
高阶贡献率（%）	35.1	42.8	49.3	56.9	60.3	73.4	76.7	85.3	92.4	92.6	94.3

表1

图8

表2

图9

图10

图11

参考文献 23

1	Markowitz H. Portfolio selection［J］. Journal of Finance， 1952， 7（7）： 77-91.
2	Sharpe W. A simplified model for portfolio analysis［J］. Management Science， 1963， 9（2）： 277-293.
3	Sharpe W. Capital asset prices： A theory of market equilibrium under conditions of risk［J］. Journal of Finance， 1964， 19（3）： 425-442.
4	Ross S. The arbitrage theory of capital asset pricing［J］. Journal of Economic Theory. 1976， 162（13）： 341-360.
5	Greenberg D， Babu A， Ang A. Factors to assets： Mapping factor exposures to asset allocations［J］. Journal of Portfolio Management， 2016， 42（5）： 18-27.
6	李向阳，高志强，田森平，等. 高阶不确定系统的线性串级自抗扰控制［J］. 控制理论与应用，2023，40（5）： 856-864.
	Li X Y， Gao Z Q， Tian S P， et al. Linear cascade disturbance rejection control for higher order system with uncertainties［J］. Control Theory & Applications， 2023，40（5）： 856-864.
7	Bouchaud J， Potters M. Theory of financial risk and derivative pricing： From statistical physics to risk management［M］. Cambridge： Cambridge university press， 2003.
8	于孝建，王秀花，徐维军. 基于滚动经济回撤约束和下半方差的最优投资组合策略［J］. 系统工程理论与实践， 2018， 38（3）： 545-555.
	Yu X J， Wang X H， Xu W J. Optimal portfolio strategy under rolling economic drawdown constraints with lower semi-variance［J］. Systems Engineering-Theory & Practice， 2018， 38（3）： 545-555.
9	Usta I， Kantar Y. Mean-variance-skewness-entropy measures： A multi-objective approach for portfolio selection［J］. Entropy， 2011， 13（1）： 117-133.
10	莫东序，郑田丹. 基于复杂网络的投资组合优化研究［J］. 中国管理科学， 2021， 29（5）： 25-33.
	Mo D X， Zheng T D. Research on portfolio optimization based on complex network［J］. Chinese Journal of Management Science， 2021， 29（5）： 25-33.
11	陈粘，黄迅，严晓凤. 结构突变下的高维汇率资产时变投资组合预测研究［J］. 中国管理科学， 2021， 29（11）： 13-22.
	Chen Z， Huang X， Yan X F. Dynamic portfolio forecasting of high-dimensions exchange rate assets under structure break［J］. Chinese Journal of Management Science， 2021， 29（11）： 13-22.
12	Tayalı S. A novel backtesting methodology for clustering in mean-variance portfolio optimization［J］. Knowledge-Based Systems， 2020， 209： 106454.
13	赵志冲，严丽霞. 多维数据下小企业违约风险过程性评价研究［J］. 中国管理科学， 2023， 31（5）： 84-92.
	Zhao Z C， Yan L X. Research on process evaluation of default risk of small enterprisesunder multidimensional data［J］. Chinese Journal of Management Science， 2023， 31（5）： 84-92.
14	Bollerslev T， Engle R， Wooldridge J. A capital asset pricing model with time-varying covariances［J］. The Journal of Political Economy， 1998， 96（1）： 116-131.
15	Hoerl A， Kennard R. Ridge regression： Biased estimation for non-orthogonal problems［J］. Techomerics， 1970， 12（1）： 55-67.
16	严杰，陈文，程鑫，等. 关于企业供应链风险管理研究的探讨［J］. 物流工程与管理， 2017， 39（4）： 105-107.
	Yan J， Chen W， Cheng X， et al. Discussion on enterprise supply chain risk management［J］. Logistics Engineering and Management， 2017， 39（4）： 105-107.
17	程国平，邱映贵. 供应链风险传导模式研究［J］. 武汉理工大学学报：社会科学版， 2009， 22（2）： 36-41.
	Cheng G P， Qiu Y G. Research on the mode of the supply chain risk conduction［J］. Journal of Wuhan University of Technology （Social Sciences Edition）， 2009， 22（2）： 36-41.
18	Zhou W X， Sornette D. Is there a real-estate bubble in the US？［J］. Physica A： Statistical Mechanics and its Applications， 2006， 361（1）： 297-308.
19	Battiston S， Gatti D， Gallegati M， et al. Credit chains and bankruptcy propagation in production networks［J］. Journal of Economic Dynamics and Control， 2007， 31（6）： 2061-2084.
20	李冰清，张潇元. 基于网络结构的企业集团内部风险传染机制研究［J］. 中国管理科学， 2023， 31（5）： 20-28.
	Li B Q， Zhang X Y. Study on the internal risk contagion mechanism of enterprise groups based on network structure［J］. Chinese Journal of Management Science， 2023， 31（5）： 20-28.
21	Ma Q T， He J M， Li S W. Endogenous network of firms and systemic risk［J］. Physica A： Statistical Mechanics and its Applications， 2018， 492（3）： 2273-2280.
22	Reuter C， Foerstl K， Hartmann E， et al. Sustainable global supplier management： The role of dynamic capabilities in achieving competitive advantage［J］. Journal of Supply Chain Management， 2010， 46（2）： 45-63.
23	钱茜，徐凯. 企业家社交关系对关联信用风险传染影响的研究——基于双层网络视角［J］. 中国管理科学， 2020， 28（11）： 35-42.
	Qian Q， Xu K. Research on the influence of entrepreneur’s social relations on associated credit risk contagion：An duallayer network perspective［J］. Chinese Journal of Management Science， 2020， 28（11）： 35-42.

试验	模型	风险	自抗扰	收益	夏普
训练	MV	0.00001	1/0.332	0.00136	0.414
训练	PRADR	0.00051	1/0.237	0.00135	0.059
测试	MV	0.00009	1/0.332	-0.00139	-0.146
测试	PRADR	0.00033	1/0.237	0.00450	0.248