基于动态因子Copula模型的行业间系统性风险分析

doi:10.16381/j.cnki.issn1003-207x.2018.03.001

摘要/Abstract

摘要： 国际金融市场间的相关关系以及系统性风险受到很多学者的重视，本文则以我国股市的行业指数作为研究对象进行实证研究。通过构建动态因子Copula模型，文章对行业的日收益率数据进行了动态相关性分析，并基于风险预期占比度量了我国行业之间系统性风险的溢出效应。本文分析了2006年1月4日至2016年7月1日的28个行业指数数据，基于GAS动态负荷因子的变化路径来刻画其相关关系，通过风险预期占比来研究行业间的风险溢出效应。研究表明，各个行业指数收益率之间存在较强的关联性。就单个行业来说，化工行业与其他行业关系最为不稳定。就金融与非金融行业而言，金融行业对非金融行业的影响较大且较为平稳。本文所得研究结果可以为投资者和风险管理者在进行决策时提供一定的指导。

关键词: 系统性风险, 因子copula, GAS模型, 溢出效应

Abstract: The correlation among international financial markets and the research on systemic risk have been emphasized by scholars. This paper focuses on the domestic stock sector indexes for empirical analysis. In this article, a dynamic factor Copula model is constructed to analyze the dynamic relationship of the daily returns, and further to measure the spillover effects of systemic risks among industries based on EPR (Expected proportion of industries at risk). For empirical research, this paper chooses intra-day stock prices of respective 28 industries, ranging from January 4^th 2006 to July 1^st 2016. Based on the path of dynamic factor loading or EPR, the correlations and risk spillover effects between industries can be discussed. Studies show that close correlations exist between returns of sector indexes. In terms of a single trade, the stock index of chemical industry is most susceptible to other industries.In terms of financial and non-financial sector, it is suggested that the financial sectors have a great and relatively stable influence on the non-financial ones. These results could provide investors and risk managers with guidance in decisions-making.

Key words: systemic risks, factor copula, GAS model, spillover effects

中图分类号:

C934

叶五一, 谭轲祺, 缪柏其. 基于动态因子Copula模型的行业间系统性风险分析[J]. 中国管理科学, 2018, 26(3): 1-12.

YE Wu-yi, TAN Ke-qi, MIAO Bai-qi. Analysis of Systemic Risk among Industries via Dynamic Factor Copulas[J]. Chinese Journal of Management Science, 2018, 26(3): 1-12.

参考文献

[1] 张晓朴. 系统性金融风险研究:演进、成因与监管[J]. 金融监管, 2010,(7):58-67.

[2] 陈九生, 周孝华,. 基于单因子MSV-CoVaR模型的金融市场风险溢出度量研究[J]. 中国管理科学, 2017,25(1):21-26.

[3] Eross A, Urquhart A, Wolfe S. Liquidity risk contagion in the interbank market[J]. Journal of International Financial Markets, Institutions and Money, 2016,45:142-155.

[4] Schwaab B.New quantitative measures of systemic risk[M]. Financial Stability Review Special Feature E, European Central Bank, 2010:147-153.

[5] Engle R. Dynamic conditional correlation:A simple class of multivariate generalized autoregressive conditional heteroskedasticity models[J]. Journals of Business & Economic Statistics, 2002,20(3):339-350.

[6] Adrian T, Brunnermeier M K. CoVaR Staff Report 348[J]. Federal Reserve Bank of New York, 2009.

[7] Acharya V, Pedersen L H, Philippon T, et al. Measuring systemic risk[J]. The Review of Financial Studies, 2017, 30(1):2-47.

[8] 宋清华,姜玉东. 中国上市银行系统性风险度量[J]. 财经理论与实践,2014,35(192):1-3.

[9] 范小云,王道平,方意. 我国金融机构的系统性风险贡献测度与监管——基于边际风险贡献与杠杆率的研究[J]. 南开经济研究, 2011,(4):3-20.

[10] 周天芸,周开国,黄亮. 机构集聚、风险传染与香港银行的系统性风险[J]. 国际金融研究, 2012,(4):77-87.

[11] 沈悦,戴士伟,罗希. 中国金融业系统性风险溢出效应测度基于GARCH-Copula-CoVaR模型的研究[J]. 当代经济科学, 2014,36(6):30-38.

[12] 白雪梅,石大龙. 中国金融体系的系统性风险度量[J]. 国际金融研究, 2014,(6):75-85.

[13] 曾裕峰, 简志宏, 彭伟. 中国金融业不同板块间风险传导的非对称性研究——基于非对称MVMQ-CAViaR模型的实证分析[J]. 中国管理科学, 2017,25(8):58-67.

[14] Bradley B O, Taqqu M S. Financial risk and heavy tails[K]. In:Handbook of Heavy Tailed Distributions in Finance. Elsevier, 2003:27-33.

[15] Dong H O, Patton A J. Time-varying systemic risk:Evidence from a dynamic copula model of CDS spreads[J]. Journal of Business & Economic Statistics, 2017:1-15.

[16] Creal D D, Koopman S J, Lucas A. Generalized autoregressive score models with applications[J]. Journal of Applied Econometrics, 2013, 28(5):777-795.

[17] 陈田,秦学志. 债务抵押债券CDO定价模型研究综述[J]. 管理学报, 2008, 5(4):616-624.

[18] Wu P C, Kao L J, Lee C F. Factor copula for defaultable basket credit derivatives[M]. New York:Springer New York, 2015:639-655.

[19] Patton A J. A review of Copula models for economic time series[J]. Journal of Multivariate Analysis, 2012, 110(2012):4-18.

[20] Patton A J. Modeling asymmetric exchange rate dependence[J]. International Economic Review, 2006, 2(47):527-556.

[21] Silva Filho O C, Ziegelmann F A. Assessing some stylized facts about financial market indexes:A Markov Copula approach[J].Journal of Economic Studies, 2014, 41(2):253-271.

[22] 陈志平,宋振霞. Copula函数在多因子模型系数估计中的应用[J]. 系统工程理论与实践, 2013, 33(10):2471-2473.

[23] Dong H O, Patton A J. Modelling dependence in high dimensions with factor copulas[R]. working paper, Duke University, 2012:4-5.

[24] Patton A J, Irving D L S. Dynamic Copula models and high frequency data[J]. Journal of Empirical Finance, 2015, 30(1):120-135.

[25] Blasques F, Koopman S J, Lucas A. Specified generalized autoregressive score models:Feedback effects, contraction conditions and asymptotic properties[R]. Tinbergen Institute Discussion Paper, 2014, 74(3).

[26] Dong H O, Patton A J. Time-varying systemic risk:Evidence from a dynamic copula model of CDS spreads[R]. Working paper, Economic Research Initiatives at Duke, 2013:28-30.

[27] Engle R F, Kelly B. Dynamic equicorrelation[J]. Journal of Business and Economic Statistics, 2012, 30(2):212-228.

[28] Engle R F. Autoregressive Conditional Heteroskedasticity with Estimates of the Variance of UK Inflation[J]. Econometrica, 1982, 50(4):987-1007.

[29] Bollerslev T. Generalized autoregressive conditional Heteroskedasticity[J], Journal of Econometrics, 1986, 31(3):307-327.