Skip to content
KPI Science News
ISSN 2617-5509 · e‑ISSN 2663-7472 Open Access
Vol. · Issue 3 · 2021 Dec 15, 2021 article

PROBABILISTIC MODELLING OF OPERATIONAL RISKS

LL
Liudmyla B. Levenchuk Corresponding Igor Sikorsky Kyiv Polytechnic Institute, Ukraine
VH
Vira G. Huskova Igor Sikorsky Kyiv Polytechnic Institute, Ukraine
PB
Petro I. Bidyuk Igor Sikorsky Kyiv Polytechnic Institute, Ukraine
Pages26-37 PublishedDec 15, 2021
PDF (Ukrainian)
KPISN 3 VOL · 3
VOL · ISS 3 · 2021 No. 3 (2021) View issue

Abstract

Background . Operational risks are inherent to all types of human activity, including industrial production, transport, financial services, etc. Risks of this kind are characterized by many uncertainties, incompleteness and low data quality, that complicate prediction and assessment. To perform modelling of the process associated with operational risk it is necessary to carry out a proper data processing as well as identification, taking into consideration possible uncertainties. The probabilistic approach to modelling is very helpful in solving the problems. Objective . The purpose of the paper is to make brief overview of probabilistic data analysis methods designed to build mathematical models of operational risks. To develop a new probabilistic model in the form of a Bayesian network to describe formally the operational risk of fraud associated with actuarial processes. Methods . The basic methodology used for data and expert estimates processing are Bayesian data analysis techniques that help to construct probabilistic models in the form of Bayesian networks. Results . The proposed modelling method was applied to constructing model of operational risk, more specifically risk of fraud in actuarial sphere. To construct the model the problem was analysed, a set of variables was selected, and prior estimates for conditional probabilities were estimated. The final model was constructed using the modelling system GeNIe. The model functioning was demonstrated using illustrative example. Conclusions . It was shown that modelling, estimating and forecasting financial and other types of risks is important practical problem that can be solved using probabilistic approach, namely Bayesian methodology that helps to identify and take into consideration possible uncertainties of data and expert estimates. The operational risk model constructed using the methodology illustrates the possibilities of app.

Keywords

#operational risks #probabilistic modelling #Bayesian methodology #expert estimates #actuarial processes

References

  1. Basel Committee on Banking Supervision. International Convergence of Capital Measurement and Capital Standards. A Revised Framework. Comprehensive Version, Basel, Switzerland: Bank for International Settlements, 2006.
  2. M. G. Cruz, Modeling, Measuring and Hedging Operational Risk, London: Wiley, 2002.
  3. C. Alexander, Operational Risk: Regulation, Analysis and Management, Harlow, England: Pearson Education Ltd, 2003.
  4. P. V. Shevchenkо, Modeling Operational Risk Using Bayesian Inference, New York: Springer, 2011.
  5. S. O. Dmytrov, Modeling of operational risk assessment of a commercial bank, Sumy, Ukraine: SHEI “Ukrainian Academy of Banking”, 2010.
  6. A. Frachot, P. Georges, and T. Roncalli, “Loss distribution approach for operational risk,” SSRN Electron. J., pp. 56–99, 2001. doi: 10.2139/ssrn.1032523.
  7. С. Alexander, “Bayesian methods for measuring operational risk,” SSRN Electron. J., pp. 45–65, 2000. doi: 10.2139/ssrn.248148.
  8. Х. Hao, “Operational risk control of commercial banks based on Bayesian network,” Proc. 2013 ICETIS, 2013, pp. 913–918. doi: 10.2991/icetis-13.2013.209.
  9. Yongqi Yang, Wanlei Xue, Chaoyuan Li, Xin Zhao, and Nan Xu, “An AHP-Bayesian model for operational risk evaluation of integrated energy systems,” IOP Conf. Ser. Materials Sci. Eng., vol. 484, pp. 1–8, 2019. doi: 10.1088/1757-899X/484/1/012052.
  10. Y. K. Yoon, “Modeling operational risk in financial institutions using Bayesian networks”, J. Risk Insur., UK, №74(4), pp. 795–827, 2007.
  11. M. Neil, N. Fenton, and M. Tailor, “Using Bayesian networks to model expected and unexpected operational losses,” Risk Analysis, vol. 25, no. 4, pp. 963–972, Aug. 2005. doi: 10.1111/j.1539-6924.2005.00641.x.
  12. A. Stuart and K. Ord, Kendalls Advanced Theory of Statistics, vol. 1, Wiley, 1994.
  13. S. Figini, L. Gao and P. Giudici, “Bayesian operational risk models,” J. Oper. Risk, vol. 10, no. 2, pp. 45–60, 2015. doi: 10.21314/JOP.2015.155
  14. C. K. Chow and C. N. Liu, “Approximating discrete probability distributions with dependence trees,” IEE Trans. Inf. Theory, vol. 14, pp. 462–467, 1968.
  15. T. Minka et al. (2014). Infer.NET 2.6 Microsoft Research Cambridge [Online]. Available: http://research.microsoft.com/infernet
  16. M. Neil, D. Hager and L. B. Andersen, “Modeling operational risk in financial institutions using hybrid,” J. Oper. Risk, vol. 4, no. 1, pp. 1–28, 2009.
§ 06 — Related

Similar articles in this journal

Related peer-reviewed studies published in this journal.
View all issues

MODIFICATION OF THE SEPARABLE PHYSICS-INFORMED NEURAL NETWORK METHOD FOR THE APPROXIMATE SOLUTION OF A MIXED PROBLEM FOR THE HEAT CONDUCTION EQUATION IN AN ELLIPTICAL DOMAIN

Valerii Los, Oleksandr Riakhin

Background. Parabolic differential equations are mathematical models in engineering designed to study wide range of applied models. This necessitates the development of…

#heat conduction equation #mixed parabolic problem #elliptical domain #physics-informed neural network
10.20535/kpisn.2026.3.367316

Most read articles by the same author(s)

Similar Articles

1-10 of 32

You may also start an advanced similarity search for this article.