吉林大学学报(工学版) ›› 2020, Vol. 50 ›› Issue (5): 1862-1869.doi: 10.13229/j.cnki.jdxbgxb20190507

• 计算机科学与技术 • 上一篇    

基于贝叶斯模型与机器学习算法的金融风险网络评估模型

李阳1(),李硕2(),井丽巍3   

  1. 1.吉林财经大学 会计学院,长春 130117
    2.吉林财经大学 公共管理学院,长春 130117
    3.吉林省科学技术信息研究所,长春 130021
  • 收稿日期:2019-05-20 出版日期:2020-09-01 发布日期:2020-09-16
  • 通讯作者: 李硕 E-mail:9106283@qq.com;ls2925@163.com
  • 作者简介:李阳(1980-),女,副教授,博士.研究方向:人工智能,金融学与管理学.E-mail:9106283@qq.com
  • 基金资助:
    吉林省自然科学基金项目(20190201134JC);吉林省哲学社会科学基金项目(2018B93);国家留学基金委项目(CSC-201902485002)

Estimate model based on Bayesian model and machine learning algorithms applicated in financial risk assessment

Yang LI1(),Shuo LI2(),Li-wei JING3   

  1. 1.College of Accounting, Jilin University of Finance and Economics, Changchun 130117,China
    2.School of Public Administration, Jilin University of Finance and Economics, Changchun 130117, China
    3.Jilin Province Institute of Science and Technology Information, Changchun 130021, China
  • Received:2019-05-20 Online:2020-09-01 Published:2020-09-16
  • Contact: Shuo LI E-mail:9106283@qq.com;ls2925@163.com

摘要:

本文以贝叶斯方法为基础构建了用于估计银行间负债的模型,并利用机器学习算法构造了可在条件分布基础上进行抽样的Gibbs取样器,抽样被用于压力测试,以此给出所有可能测试结果的概率。最后,推导出了银行的违约概率并讨论其对包含在网络模型中的先验信息的敏感性,帮助金融监管部门评估金融机构的违约风险,减少系统性金融风险,维护金融市场的稳定。

关键词: 计算机应用, 金融风险评估, 贝叶斯模型, 机器学习算法, Gibbs取样器

Abstract:

A model to estimate the inter-bank liabilities was construct based on the Bayesian method, and then use machine learning algorithms to construct a Gibbs sampler to sample on the basis of conditional distribution. The sampling is used for stress testing and give the probability of all possible test results. Finally, as a model application, this paper will derive the bank's default probability and discuss its sensitivity to a priori information contained in the network model, helping financial regulators assess financial institutions' default risk,reduce systemic financial risks and maintainfinancial market stability.

Key words: computer application, financial risk assessment, Bayesian model, machine learning algorithm, Gibbs sampler

中图分类号: 

  • TP391

(a)

银行1"

(b)

银行2"

(c)

银行3"

1 刘颖, 张丽娟, 韩亚男,等. 基于粒子群协同优化算法的供应链金融信用风险评价模型[J]. 吉林大学学报:理学版, 2018, 56(1): 119-125.
Liu Ying, Zhang Li-juan, Han Ya-nan,et al. Financial credit risk evaluation model of supply chain finance based on particle swarm cooperative optimization algorithm[J]. Journal of Jilin University (Science Edition), 2018, 56(1): 119-125.
2 王妍,陈守东.尾部极值分布下的系统性金融风险度量及影响因素分析[J]. 数理统计与管理,2014(6):1010-1020.
Wang Yan,Chen Shou-dong. Measuring systemic financial risk and analyzing the influence factors: under the assumption of extreme distribution for the tail[J]. Jouranl of Applied Statistics and Management, 2014(6):1010-1020.
3 王可, 王德辉. 双参数广义几何分布的Bayes估计及其应用[J]. 吉林大学学报:理学版, 2019, 57(5): 1104-1110.
Wang Ke, Wang De-hui. Bayesian estimation for generalized geometric distribution with double-parameters and its application[J]. Journal of Jilin University(Science Edition), 2019, 57(5): 1104-1110.
4 赵进文, 张胜保, 韦文彬. 系统性金融风险度量方法的比较与应用[J]. 统计研究, 2013, 30(10):46-53.
Zhao Jin-wen,Zhang Sheng-bao,Wei Wen-bin. A Comparison and application of systemic risk measures[J]. Statistical Research, 2013, 30(10):46-53.
5 郝志峰,王日宇,蔡瑞初,等. 基于贝叶斯网络与语义树的隐私数据发布方法[J]. 计算机工程, 2019, 45(4): 124-129.
Hao Zhi-feng,Wang Ri-yu,Cai Rui-chu,et al. Privacy data publishing method based on Bayesian network and semantic tree[J]. Computer Engineering, 2019, 45(4): 124-129.
6 许硕, 唐作其, 王鑫. 基于D-AHP与灰色理论的信息安全风险评估[J]. 计算机工程, 2019, 45(7): 194-202.
Xu Shuo, Tang Zuo-qi, Wang Xin. Information security risk assessment based on D-AHP and grey theory[J]. Computer Engineering, 2019, 45(7): 194-202.
7 Amini H, Cont R, Minca A. Resilience to Contagion in Financial Networks[J]. Mathematical Finance, 2016, 26(2): 1-40.
8 Barvinok A. Matrices with prescribed row and column sums[J]. Linear Algebra and its Applications2012, 436: 820-844.
9 Stock J H, Watson M W.New indexes of coincident and leading economic indicators[J].NBER Macroeconomics Annual, 1989, 4: 351-409.
10 Shin H S. Financial intermediation and the post-crisis financial system[J]. Bis Working Papers, 2010, 68(304): 1-18.
11 Nakajima J,West M.Dynamic factor volatility modeling:A Bayesian latent threshold approach[J].Journal of Financial Econometrics, 2013, 11(1): 116-153.
12 Kenourgios D,Samitas A,Paltalidis N.Financial crises and stock market contagion in a multivariate time-varying asymmetric framework[J].Journal of International Financial Markets,Institutions and Money,2011, 21(1): 92-106.
13 Erdős, Pál, Rényi, et al. On random graphs I[J]. Publicationes Mathematicae, 1959, 4: 3286-3291.
14 王辉, 娄亚龙, 戴田旺, 等. 基于BNAG模型的脆弱性评估算法[J]. 计算机工程, 2019, 45(9): 128-135, 142.
Wang Hui, Lou Ya-long, Dai Tian-wang, et al. Vulnerability evaluation algorithm based on BNAG model[J]. Computer Engineering, 2019, 45(9): 128-135, 142.
[1] 周炳海,何朝旭. 基于线边集成超市的混流装配线动态物料配送调度[J]. 吉林大学学报(工学版), 2020, 50(5): 1809-1817.
[2] 蒋磊,管仁初. 基于多目标进化算法的人才质量模糊综合评价系统设计[J]. 吉林大学学报(工学版), 2020, 50(5): 1856-1861.
[3] 赵宏伟,刘晓涵,张媛,范丽丽,龙曼丽,臧雪柏. 基于关键点注意力和通道注意力的服装分类算法[J]. 吉林大学学报(工学版), 2020, 50(5): 1765-1770.
[4] 管乃彦,郭娟利. 基于姿态估计算法的组件感知自适应模型[J]. 吉林大学学报(工学版), 2020, 50(5): 1850-1855.
[5] 车翔玖,董有政. 基于多尺度信息融合的图像识别改进算法[J]. 吉林大学学报(工学版), 2020, 50(5): 1747-1754.
[6] 刘洲洲,尹文晓,张倩昀,彭寒. 基于离散优化算法和机器学习的传感云入侵检测[J]. 吉林大学学报(工学版), 2020, 50(2): 692-702.
[7] 王晓辉,吴禄慎,陈华伟. 基于法向量距离分类的散乱点云数据去噪[J]. 吉林大学学报(工学版), 2020, 50(1): 278-288.
[8] 张笑东,夏筱筠,吕海峰,公绪超,廉梦佳. 大数据网络并行计算环境中生理数据流动态负载均衡[J]. 吉林大学学报(工学版), 2020, 50(1): 247-254.
[9] 陈蔓,钟勇,李振东. 隐低秩结合低秩表示的多聚焦图像融合[J]. 吉林大学学报(工学版), 2020, 50(1): 297-305.
[10] 金顺福,郄修尘,武海星,霍占强. 基于新型休眠模式的云虚拟机分簇调度策略及性能优化[J]. 吉林大学学报(工学版), 2020, 50(1): 237-246.
[11] 邓钧忆,刘衍珩,冯时,赵荣村,王健. 基于GSPN的Ad⁃hoc网络性能和安全平衡[J]. 吉林大学学报(工学版), 2020, 50(1): 255-261.
[12] 王铁君,王维兰. 基于本体的唐卡图像标注方法[J]. 吉林大学学报(工学版), 2020, 50(1): 289-296.
[13] 李雄飞,王婧,张小利,范铁虎. 基于SVM和窗口梯度的多焦距图像融合方法[J]. 吉林大学学报(工学版), 2020, 50(1): 227-236.
[14] 王洪雁,邱贺磊,郑佳,裴炳南. 光照变化下基于低秩稀疏表示的视觉跟踪方法[J]. 吉林大学学报(工学版), 2020, 50(1): 268-277.
[15] 周柚,杨森,李大琳,吴春国,王岩,王康平. 基于现场可编程门电路的人脸检测识别加速平台[J]. 吉林大学学报(工学版), 2019, 49(6): 2051-2057.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!