Systemic Risk and Bank Networks: A Use of Knowledge Graph with ChatGPT

Author:

Lyu Ren-Yuan1,Chen Ren-Raw2,Chung San-Lin13,Zhou Yilu4

Affiliation:

1. College of Management, Chang Gung University, 259 Wenhua First Road, Taoyuan 33302, Taiwan

2. Gabelli School of Business, Fordham University, 45 Columbus Avenue, Room 609, New York, NY 10019, USA

3. College of Management, National Taiwan University, 1 Section 4 Roosevelt Road, Taipei 106319, Taiwan

4. Gabelli School of Business, Fordham University, 140 W. 62nd Street, Room 407, New York, NY 10023, USA

Abstract

In this paper, we study the networks of financial institutions using textual data (i.e., news). We draw knowledge graphs after the textual data has been processed via various natural language processing and embedding methods, including use of the most recent version of ChatGPT (via OpenAI api). Our final graphs represent bank networks and further shed light on the systemic risk of the financial institutions. Financial news reflects live how financial institutions are connected, via graphs which provide information on conditional dependencies among the financial institutions. Our results show that in the year 2016, the chosen 22 top U.S. financial firms are not closely connected and, hence, present no systemic risk.

Funder

San-Lin Chung acknowledges the financial support from the National Science and Technology Council, Taiwan

Publisher

MDPI AG

Reference41 articles.

1. From Liquidity Risk to Systemic Risk: A Use of Knowledge Graph;Chen;J. Financ. Stab.,2024

2. Denev, A. (2015). Probabilistic Graphical Models: A New Way of Thinking in Financial Modeling, Risk.

3. Investor Networks in the Stock Market;Ozsoylev;Rev. Financ. Stud.,2014

4. Investor network: Implications for information diffusion and asset prices;Chung;Pac.-Basin Financ. J.,2018

5. Network topology of the interbank market;Boss;Quant. Financ.,2004

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3