基于动态CRITIC赋权的中国金融压力指数构建与金融风险识别

doi:10.16716/j.cnki.65-1030/f.2024.01.003

摘要/Abstract

摘要：

合理测度和准确识别金融市场风险，对于稳定经济、有效防范金融风险意义重大。依据中国金融市场特征，本文采用动态CRITIC法计算权重并建立金融压力指数测度模型，通过非参数统计核密度法和B-N数据分解法对金融压力指数的分布和金融风险状态进行估计与识别，采用马尔科夫区制转换模型检验高低风险转换概率，并与基于确定项进行识别的结果进行对比分析。研究表明：基于动态CRITIC赋权的金融压力指数更能反映金融市场的极端值和金融风险，随机冲击是影响我国金融压力指数的重要因素，两种金融风险识别方法均可识别金融风险状态且各有优势，研究中可以互为补充。今后应建立更加完善的风险测度指标体系，进一步完善监管制度，加强对金融风险的宏观审慎管理和防范。

关键词: 金融压力指数, 动态CRITIC法, B-N数据分解法, 马尔科夫区制转换模型

Abstract:

Reasonable measurement and accurate identification of financial market risks are of great significance for stabilizing the economy and effectively preventing financial risks. According to the characteristics of China's financial market, this paper uses the dynamic CRITIC method to establish the measurement model to calculate the financial stress index, and estimates and identifies the distribution and risk state of financial stress index through non-parametric statistical kernel density method and B-N data decomposition method and the Markov Regime Switching Model is used to test the high and low risk transformation probability of financial stress index to compare with the results identified by deterministic items. The research results indicate that the financial pressure index based on dynamic CRITIC weighting method can better reflect the extreme values and financial risks of the financial market. Random shocks are an important factor affecting China's financial pressure index, and both financial risk identification methods can identify the financial risk status and each has its own advantages. In the future, it is of great necessity to establish a more complete risk measurement index system, further improve the regulatory system, and strengthen macro-prudential management and prevention of financial risks.

Key words: financial stress index, dynamic CRITIC method, B-N data decomposition method, Markov regime switching model

中图分类号:

F832

祝志川, 蒋犇. 基于动态CRITIC赋权的中国金融压力指数构建与金融风险识别[J]. 新疆财经, 2024, (1): 21-33.

ZHU Zhichuan, JIANG Ben. Construction and Application of China's Financial Stress Index Based on Dynamic CRITIC Weighting Method and Financial Risk Identification[J]. Finance & Economics of Xinjiang, 2024, (1): 21-33.

图/表 12

表1

表2

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

参考文献 26

[1]	ILLING M, LIU Y. Measuring financial stress in a developed country:an application to Canada[J]. Journal of financial stability, 2006(3):243-265.
[2]	HAKKIO C, KEETON W R. Financial stress:what is it,how can it be measured,and why does it matter?[J]. Economic review, 2009(2):5-50.
[3]	BALAKRISHNAN R, DANNINGER S, ELEKDAG S, et al. The transmission of financial stress from advanced to emerging economies[J]. Emering markets finance and trade, 2011(sup2):40-68.
[4]	OZCELEBI O. Assessing the impacts of financial stress index of developed countries on the exchange market pressure index of emerging countries[J]. International review of economics and finance, 2020(9):288-302.
[5]	CEVIK E, DIBOOGLU S, KENC T. Financial stress and economic activity in some emerging Asian economies[J]. Research in international business and finance, 2016(C):127-139.
[6]	王维国, 王际皓. 货币、银行与资产市场风险状况的识别:基于金融压力指数与MSIH-VAR模型的实证研究[J]. 国际金融研究, 2016(8):71-81.
[7]	MACDONALD R, SOGIAKAS V, TSOPANAKIS A. Volatility co-movements and spillover effects within the Eurozone economies:a multivariate GARCH approach using the financial stress index[J]. Journal of international financial markets,institutions and money, 2018(1):17-36.
[8]	丁岚, 李鹏涛, 刘立新. 中国金融压力指数的构建与应用[J]. 统计与信息论坛, 2019(10):63-72.
[9]	陈忠阳, 许悦. 我国金融压力指数的构建与应用研究[J]. 当代经济科学, 2016(1):27-35+125.
[10]	丁述军, 庄须娟, 李文君. 区域金融风险部门间传染机理与实证分析[J]. 经济经纬, 2019(3):1-8.
[11]	姚晓阳, 孙晓蕾, 李建平. 考虑市场相关性的中国金融压力指数构建方法与实证[J]. 管理评论, 2019(4):34-41.
[12]	秦建文, 王涛. 中国金融压力与宏观经济动态效应研究:基于MS-VAR模型的实证分析[J]. 中央财经大学学报, 2017(9):32-42.
[13]	郑金麟, 张业圳, 谢八妹. 我国系统性金融风险的衡量与识别[J]. 金融监管研究, 2019(12):54-65.
[14]	张勇, 彭礼杰, 莫嘉浩. 中国金融压力的度量及其宏观经济的非线性效应[J]. 统计研究, 2017(1):67-79.
[15]	章曦. 中国系统性金融风险测度、识别和预测[J]. 中央财经大学学报, 2016(2):45-52.
[16]	马勇, 黄科. 金融压力指数及其政策应用:基于中国的实证分析[J]. 金融监管研究, 2019(7):1-17.
[17]	许悦. 系统性压力综合指数的有效性研究[J]. 统计与决策, 2017(2):166-170.
[18]	YAO X Y, LE W, SUN X L, et al. Financial stress dynamics in China:an interconnectedness perspective[J]. Inter-national review of economics and finance, 2020(7):217-238.
[19]	JAMES C, MORLEY. A state-space approach to calculating the Beveridge-Nelson decomposition[J]. Economics Letters, 2002(1):123-127.
[20]	李子联, 陈强. 中国对外贸易的疫情冲击效应[J]. 现代经济探讨, 2021(4):75-82.
[21]	AKLAN N A, ÇINAR M, AKAY H K. Financial stress and economic activity relationship in Turkey: post-2002 period[J]. Journal of management & economics, 2015(2):567-580.
[22]	KIM H, SHI W, KIM H H. Forecasting financial stress indices in Korea:a factor model approach[J]. Empirical economics, 2015(1):1-40.
[23]	BEVERIDGE S, NELSON C R. A new approach to decomposition of economic time series into permanent and transitory components with particular attention to measurement of the ‘business cycle’[J]. Journal of monetary economics, 1981(2):151-174.
[24]	KIM C J. Markov-switching and the Beveridge-Nelson decomposition:has US output persistence changed since 1984?[J]. Journal of econometrics, 2008(2):227-240.
[25]	清华大学国家金融研究院金融与发展研究中心课题组. 中国系统性金融压力的监测[J]. 国际金融研究, 2019(12):3-12.
[26]	李子联. 进出口贸易中不确定性事件的冲击效应[J]. 财经理论与实践, 2010(6):77-82.

子市场	子市场风险指标	符号	计算方式
债券市场	期限利差	tesp	10年期国债对数收益率与1年期国债对数收益率之差
债券市场	主权债务利差	sdim	10年期中国国债与10年期美国国债收益率差值的绝对值
外汇市场	美元兑人民币	uscn	美元兑人民币汇率的变动率
外汇市场	欧元兑人民币	eucn	欧元兑人民币汇率的变动率
股票市场	上证指数收益率	shcy	上证指数对数收益率的GARCH（1，1）时变方差
	深证成指收益率	szcy	深证成指对数收益率的GARCH（1，1）时变方差
	沪深300指数收益率	hscy	沪深300指数对数收益率的GARCH（1，1）时变方差
外部市场	标普500指数波动率	spy	标普500指数的GARCH（1，1）时变方差
	WTI原油期货的CMAX	wti	WTI原油期货收盘价一年滚动窗口期内的最大累积损失
	黄金期货收益率	zgtl	黄金期货股指收益率的对数收益率
	美元指数收益率	dlhl	美元指数股指收益率的对数收益率

市场类型	观测值	平均风险值	标准差	方差	峰度	偏度	最小值	最大值
债券市场	3882	0.4042	0.1443	0.0208	-1.0140	0.0942	0.0572	0.7783
外汇市场	3882	0.4908	0.0419	0.0017	11.8590	-0.3385	0.0001	0.9451
股票市场	3882	0.0898	0.1070	0.0114	10.6339	2.8546	0.0013	0.9893
外部市场	3882	0.3816	0.0463	0.0210	1.0496	0.5876	0.2302	0.6049