یادگیری ماشین در تخمین سرمایه پوششی ریسک‌عملیاتی بانک‌ها با رویکرد توزیع زیان | ||
| چشم انداز مدیریت مالی | ||
| دوره 13، شماره 42، شهریور 1402، صفحه 9-34 اصل مقاله (827.29 K) | ||
| نوع مقاله: علمی - پژوهشی | ||
| شناسه دیجیتال (DOI): 10.48308/jfmp.2023.103948 | ||
| نویسندگان | ||
| مهدی اکبری1؛ احمدرضا یزدانیان* 2 | ||
| 1کارشناسی ارشد ریاضیات مالی، دانشگاه خوارزمی، تهران، ایران. | ||
| 2استادیار، گروه ریاضیات مالی، دانشگاه خوارزمی، تهران، ایران | ||
| چکیده | ||
| ریسک عملیاتی یکی از مهم‌ترین ریسک‌های مؤسسات مالی است. توجه به آن پس از مصوبات کمیته بال، در سراسر دنیا آغاز شده‌است. افزایش روزافزون زیان‌های عملیاتی در خطوط کسب‌و‌کار مختلف سبب شده‌است تا توجّه مدیران مؤسسات مالی معطوف به حوزه ریسک عملیاتی شود. در این پژوهش، روشی جهت تخمین آستانه مناسب برای داده‌های شدت زیان عملیاتی و همچنین روشی جهت طبقه‌بندی داده‌های شدت زیان عملیاتی ارائه شده‌است و سرمایه مورد نیاز برای پوشش ریسک عملیاتی با تجمیع تابع توزیع شدت و فرکانس داده‌های زیان عملیاتی و شبیه‌سازی مونتکارلو به‌دست آمده‌است. همچنین وابستگی بین سلول‌های ماتریس خطوط کسب‌و‌کار و حوادث ضررساز نیز مورد بررسی قرار گرفته‌است. برای این پژوهش داده‌های زیان عملیاتی مربوط به‌یک مجموعه بانکداری شامل چند بانک آسیایی، اروپایی و آمریکایی به‌کارگرفته شده‌است. نتایج پژوهش حاکی از آن است که رویکرد توزیع زیان با ترکیب تئوری مقدار بحرانی و الگوریتم‌های‌یادگیری ماشین (خوشه‌بندی)، همچنین رویکرد توزیع زیان با ترکیب الگوریتم‌های‌یادگیری ماشین (طبقه‌بندی)، نسبت به سایر روش‌ها کارآمدتر است. | ||
| کلیدواژه‌ها | ||
| ریسک عملیاتی؛ رویکرد توزیع زیان؛ نظریه مقدار بحرانی؛ الگوریتمهای طبقه‌بندی؛ الگوریتمهای خوشه‌بندی | ||
| عنوان مقاله [English] | ||
| Machine learning in estimating operational risk coverage capital of banks with a loss distribution Approach | ||
| نویسندگان [English] | ||
| Mahdi Akbari1؛ Ahmadreza Yazdanian2 | ||
| 1M.A. Student in Financial Mathematics, University of Kharazmi, Tehran, Iran | ||
| 2Assistant Prof., Department of Financial Mathematics, Kharazmi University, Tehran, Iran | ||
| چکیده [English] | ||
| Operational risk is one of the most important risks of financial institutions. After the approval of the Basel committee, it has started to be noticed all over the world. The increasing increase in operating losses in various business lines has caused the attention of financial institution managers to be directed to the area of operational risk. In this research, a method for estimating the appropriate threshold for operational loss data and also a method for classifying operational loss data is presented, and the capital required to cover operational risk by combining the intensity distribution function and frequency of operating loss data and Monte Carlo simulation are obtained. Also, the dependence between matrix cells of business lines and loss making events has also been investigated. For this research, operational loss data related to a banking group including several Asian, European and American banks have been used. The research results indicate that the loss distribution approach with the combination of critical value theory and machine learning algorithms (clustering), as well as the loss distribution approach with the combination of machine learning algorithms (classification), is more efficient than other methods. | ||
| کلیدواژه‌ها [English] | ||
| Operational Risk, Loss Distribution Approach, Critical Value Theory, Classification Algorithms, Clustering Algorithms | ||
| مراجع | ||
|
1.Frachot, A.,Georges, P.,Roncalli, T. (2001). Loss Distribution Approach for operational risk, National Institute of Statistics and Economic Studies (INSEE) - Center for Research in Economics andStatistics(CREST), Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1032523 2.Farhadian, A., Rostami, M., Nilchi, M. (2021). Compare Canonical Stochastic Volatility Model of focal MSGJR-GARCH to Measure the volatility of Stock Returns and Calculating VaR. Journal of Financial Management Perspective,10(32),131-158. In persian. 3.Shahriari, A.A., Karimzade, S., Bahmanesh, R. (2021). Stock portfolio optimization in fireworks algorithm using risk value and comparison with Particle Swarm Optimization (PSO). Journal of Financial Management Perspective.11(35),9-37. In persian. 4.Johnemark, A. (2012). Modeling Operational Risk.( Degree Projectin Mathematical Statistics). Royal Institute of Technology School of Engineering Sciences, Sweden. 5.Afambo,E. (2006). Operational Risk Capital Provisions for Banks and Insurance Companies(Degree of Doctor of Philosophy in the Robinson College of Business). Department of Risk Management and Insurance, Georgia State University,USA. 6.Basel Committee on banking supervision (2005). International Convergence of Capital Measurement and Capital Standards,A RevisedFramework - Comprehensive Version, Retrieved from https://www.bis.org/publ/bcbs128.htm 7.Basel Committee on banking supervision (2003). Sound Practicesfor The Management and Supervision of Operational risk. Retrieved from https://www.bis.org/publ/bcbs96.htm 8.Basel Committee on banking supervision (2001).Working Paperon The Regulatory Treatment of Operational Risk. Retrieved from https://www.bis.org/publ/bcbs_wp8.pdf 9.Basel Committee on Banking Supervision (2009). Bank for International Settlements. Results from the 2008 Loss Data Collection Exercise for Operational Risk. Retrieved from https://www.bis.org/publ/bcbs160.htm 10.Chavez,V., Embrechts,P., Neslehova,J. (2005). Quantitative models for Operational Risk. Journal of Banking and Finaance,30(10),2635-2658. 11.Chapell,A., Crama,Y., Hubner, J., Peters, J. (2004). Operational Risk:Implications for Risk Measurement and Management in the Financial Sector.NBB Working Paper, 51, National Bank of Belgium, Brussels, 17-18May2004, Financial Management. 12.Dhruv, D.T. (2019). Credit Risk Anal Ysis Using Machin Learning And Neural Networksneural Networks. (Submitted in partial fulfillment of the requirements for the degree of master of science in Mathematical Sciences) Open Access Master's Report, Michigan Technological University,USA. 13.Bakassa-Traore, D.E. (2015). Computing Value-At-Risk And Expected Shortfall in Operational Risk. ( Master of Science in the Program of Applied Mathematics). Ryerson University,Canada. 14.Demetis, D.S. (2018). Fighting money laundering with technology: A case study of Bank X in the UK. Decision Support Systems, 105, 96-107. 15.Dutta, K., Perry, J. (2007). A Tail of Tails: An Empirical Analysis of Loss Distribution Models for Estimating Operational Risk Capital. Working Papers, No. 06-13, Federal Reserve Bank of Boston, Boston, MA.July2007, Financial Management. 16.Dahlberg, E. (2015). Bayesian Inference Methods in Operational Risk. (Master of Mathematical Statistics). School of Engineering Sciences, Sweden. 17.Erfanian, A. (2006). Comparative study and implementation of operational risk measurement models in industry and Mine Bank. Sharif Scientific Quarterly. 34,116-143,In persian. 18.Eghtesad Novin Bank (2006). Asset management - debt and liquidity risk in financial institutions. Iran, Financial organizations – Management Books. 19.Fontnouvelle, D.P., Jordan, J., Rosengren, E. (2003, April 29). Using Loss Data to Quantify Operational Risk. Federal Reserve Bank of Boston. Paper SSRN,1-32. 20.Gabriella, L. (2011). Regulation and practice of managing bank’s operational risk.( PhD School in Business and Management).earningBudapest University, Magyarorszag. 21.Pakize, K., Nosrati, H. (2014). Estimation of operational risk capital reserve in banking industry with loss distribution approach. Journal of financial engineering. 20,1-26. In persian. 22.Bhatia, P. (2019). Data mining and Data Warehousing. India: University books. 23.Modarresi, N., Peymani, M., Darvishi, M. (2021). Estimation of Conditional Value at Risk under Stochastic Volatility Levy Processes for Tehran Stock Market. Journal of Financial Management Perspective. 11(34),69-94. In persian. | ||
|
آمار تعداد مشاهده مقاله: 3,567 تعداد دریافت فایل اصل مقاله: 1,840 |
||