Probabilistic Modeling of Operational Actuarial Risks

Insurance companies are functioning in conditions of uncertainties of various types and nature what results in respective financial risks. All these reasons lead to the problem of timely recognition and development of mechanisms for the risks management. To solve the problem appropriate mathematical models are developed to describe the risks, and methodologies proposed for their practical application. The sources of the insurance fraud are detected and respective risk classification is presented. It is shown that to describe mathematically the risks of this class it is appropriate to apply the models based on the mathematical statistics approach, regression type models, and fuzzy logic. For estimation of the risk of actuarial fraud in auto insurance a model is proposed in the form of Bayesian network. The model structure was estimated using expert and statistical information of insurance company with providing a possibility for detecting hidden interactions between selected variables. An algorithm was also developed for probabilistic inference on the network. The model constructed reflects the causal links between the risk factors and the insurance company losses. It can be applied for analysis of internal states of the company; analysis of external conditions characteristic for the company functioning; for determining probable reasons of company losses due to operational risks as well as for making appropriate managerial decisions.

Publication year: 
2013
Issue: 
2
УДК: 
519.766.4
С. 45–58. Іл. 5. Табл. 4. Бібліогр.: 25 назв.
References: 

1. B. Baesens et al., “Learning bayesian network classifiers for credit scoring using Markov chain Monte Carlo search”, in Proc. 16th Int. Conf. Pattern Recognition, Québec, Canada, August 2002, pp. 49—52.
2. Ormerod T. et al., “Using ethnography to design a mass detection tool (MDT) for the early discovery of insurance fraud”, in Proc. Conf. Human Factors in Computing Systems СHI-03, Ft. Lauderdale, Florida, 2003, pp. 650— 651.
3. E.B. Belhadji and G. Dionne, Development of an Expert System for the Automatic Fraud Detection of Automobile Insurance Fraud. Canada, Montreal: Ecole des Hutes Etudes Commerciales, 1997, 376 p.
4. J. Pathak et al., “A Fuzzy-based Algorithm for Auditors to Detect Elements of Fraud in Settled Insurance Claims”, Odette School of Business Admin., Working Paper no. 03-9, 17 p., 2003.
5. Enterprise Risk Management. UK: Committee of Sponsoring Organizations of the Threadway Commission, 2004, 300 p.
6. Бюро Страхових Історій [Електронний ресурс]. — Режим доступу: http://www.antiobman.ru. — Назва з екрана.
7. Fraud Act 2006 of United Kingdom: 8 November, 2006. The Parliament of the United Kingdom, The National Archives, 15.01.2007, p. 15.
8. R. Linkoln et al. (2003). An Exploration of Automobile Insurance Fraud [Online]. Avaliable: http://epublications. bond.edu.au/hsspubs/64
9. Энциклопедия финансового риск-менеджмента / Под ред. А.А. Лобанова и А.В. Чугунова. — М.: Альпина Бизнес Букс, 2007. — 731 c.
10. G.A. Holton, “Perspectives: Defining Risk”, Fin. Analysts J. CFA Inst., vol. 60, no. 6, pp. 19—25, 2004.
11. M.H. Tripp et al., “Quantifying operational risk in general insurance companies”, in Giro Working Party [Presented to the Institute of Actuaries], 22 March 2004, 137 р.
12. S. Shah (2003). Measuring Operational Risk Using Fuzzy Logic Modeling [Online]. Avaliable: http://www.irmi. com/ Expert/Articles/2003/Shah09.aspx
13. International Convergence of Capital Measurement and Capital Standards. A Revised Framework. Comprehensive Version. Switzerland, Basel: Basel Committee on Banking Supervision, Bank for Int. Settlements, 2006, 158 p.
14. Sound Practices for the Management and Supervision of Operational Risk. Switzerland, Basel: Basel Committee on Banking Supervision, Bank for Int. Settlements, 2003, 110 p.
15. E. Medova, “Extreme Value Theory (Extreme values and the measurement of operational risk)”, Operational Risk, pp. 17—36, July 2000.
16. Scenario-based AMA (2003). [Online]. Avaliable: http:// www.newyorkfed.org/newsevents/events/banking/2003/ con0529d.pdf
17. B. Dоbeli et al., “From operational risk to operational excellence”, in Advances in operational risk: Firm-wide issues for financial institutions, ed. P. Mestchian, 2nd ed. UK, London: Risk Books, Risk Water Group, 2003.
18. R. Kuhn and P. Neu, “Functional Correlation Approach to Operational Risk in Banking Organizations”, Physica A, no. 322, pp. 650—660, 2003.
19. Моделирование рисковых ситуаций в экономике и бизнесе: Учеб. пособие / A.M. Дубров, Б.А. Лагоша, Е.Ю. Хрусталев; под. ред. Б.А. Лагоши. — М.: Финансы и статистика, 2000. — 176 c.
20. Штовба С.Д. Введение в теорию нечетких множеств и нечеткую логику [Электронный ресурс]. — Режим доступа: http://matlab.exponenta.ru/fuzzylogic/bookl/index. php. — Название с эркана.
21. M. Neil et al., “Using bayesian networks to model expected and unexpected operational losses”, Risk Analysis, vol. 25, no. 4, pp. 34—57, 2005.
22. Згуровский М.З., Терентьев А.М., Бидюк П.И. Методы построения байесовских сетей на основе оценочных функций // Кибернетика и системный анализ. — 2008. — № 2. — С. 81—88.
23. Бідюк П.І., Кузнєцова Н.В., Терентьєв О.М. Система підтримки прийняття рішень для аналізу фінансових даних // Наукові вісті НТУУ “КПІ”. — 2011. — № 1. — С. 48—61.
24. G.F. Cooper, “The computational complexity of probabilistic inference using bayesian belief networks”, Artificial Intelligence, no. 42, pp. 393—405, 1990.
25. H. Guo and W. Hsu, A survey on algorithms for real-time bayesian network inference. Laboratory for Knowledge Discovery in Databases Department of Computing and Information Sciences, Kansas State University, 2002, 20 р.

References [transliteration]: 

1. B. Baesens et al., “Learning bayesian network classifiers for credit scoring using Markov chain Monte Carlo search”, in Proc. 16th Int. Conf. Pattern Recognition, Québec, Canada, August 2002, pp. 49–52.
2. Ormerod T. et al., “Using ethnography to design a mass detection tool (MDT) for the early discovery of insurance fraud”, in Proc. Conf. Human Factors in Computing Systems СHI-03, Ft. Lauderdale, Florida, 2003, pp. 650–651.
3. E.B. Belhadji and G. Dionne, Development of an Expert System for the Automatic Fraud Detection of Automobile Insurance Fraud. Canada, Montreal: Ecole des Hutes Etudes Commerciales, 1997, 376 p.
4. J. Pathak et al., “A Fuzzy-based Algorithm for Auditors to Detect Elements of Fraud in Settled Insurance Claims”, Odette School of Business Admin., Working Paper no. 03-9, 17 p., 2003.
5. Enterprise Risk Management. UK: Committee of Sponsoring Organizations of the Threadway Commission, 2004, 300 p.
6. Bi͡uro Strakhovykh Istoriĭ [Elektronnyĭ resurs]. – Rez͡hym \ dostupu: http://www.antiobman.ru. – Nazva z ekranu.
7. Fraud Act 2006 of United Kingdom: 8 November, 2006. The Parliament of the United Kingdom, The National Archives, 15.01.2007, p. 15.
8. R. Linkoln et al. (2003). An Exploration of Automobile Insurance Fraud [Online]. Avaliable: http://epublications.bond.edu.au hsspubs/64
9. Ėnt͡siklopedii͡a finansovogo risk-menedzhmenta / Pod red. A.A. Lobanova i A.V. Chugunova. – M.: Al'pina Biznes Buks, 2007. – 731 s.
10. G.A. Holton, “Perspectives: Defining Risk”, Fin. Analysts J. CFA Inst., vol. 60, no. 6, pp. 19–25, 2004.
11. M.H. Tripp et al., “Quantifying operational risk in general insurance companies”, in Giro Working Party [Presented to the Institute of Actuaries], 22 March 2004, 137 р.
12. S. Shah (2003). Measuring Operational Risk Using Fuzzy Logic Modeling [Online]. Avaliable: http://www.irmi.com/Expert Articles/2003/Shah09.aspx
13. International Convergence of Capital Measurement and Capital Standards. A Revised Framework. Comprehensive Version. Switzerland, Basel: Basel Committee on Banking Supervision, Bank for Int. Settlements, 2006, 158 p.
14. Sound Practices for the Management and Supervision of Operational Risk. Switzerland, Basel: Basel Committee on Banking Supervision, Bank for Int. Settlements, 2003, 110 p.
15. E. Medova, “Extreme Value Theory (Extreme values and the measurement of operational risk)”, Operational Risk, pp. 17–36, July 2000.
16. Scenario-based AMA (2003). [Online]. Avaliable: http:/ www.newyorkfed.org/newsevents/events/banking/2003/con0529d.pdf
17. B. Döbeli et al., “From operational risk to operational excellence”, in Advances in operational risk: Firm-wide issues for financial institutions, ed. P. Mestchian, 2nd ed. UK, London: Risk Books, Risk Water Group, 2003.
18. R. Kuhn and P. Neu, “Functional Correlation Approach to Operational Risk in Banking Organizations”, Physica A, no. 322, pp. 650–660, 2003.
19. Modelirovanie riskovykh situat͡siĭ v ėkonomike i biznese: Ucheb. posobie / A.M. Dubrov, B.A. Lagosha, E.I͡U. Khrustalev; pod. red. B.A. Lagoshi. – M.: Finansy i statistika, 2000. – 176 s.
20. Shtovba S.D. Vvedenie v teorii͡u nechetkikh mnozhestv i nechetkui͡u logiku [Ėlektronnyĭ resurs]. – Rezhim dostupa: http:/ matlab.exponenta.ru/fuzzylogic/bookl/index.php. – Nazvanie s ėrkana.
21. M. Neil et al., “Using bayesian networks to model expected and unexpected operational losses”, Risk Analysis, vol. 25, no. 4, pp. 34–57, 2005.
22. Zgurovskiĭ M.Z., Terent'ev A.M., Bidi͡uk P.I. Metody \ postroenii͡a baĭesovskikh seteĭ na osnove ot͡senochnykh \ funkt͡siĭ // Kibernetika i sistemnyĭ analiz. – 2008. – # 2. – S. 81 88.
23. Bidi͡uk P.I., Kuzni͡et͡sova N.V., Terent′i͡ev O.M. Systema pidtrymky pryĭni͡atti͡a rishen′ dli͡a analizu finansovykh danykh // Naukovi visti NTUU “KPI”. – 2011. – # 1. – S. 48–61.
24. G.F. Cooper, The computational complexity of probabilistic inference using bayesian belief networks, Artificial Intelligence, no. 42, pp. 393–405, 1990.
25. H. Guo and W. Hsu, A survey on algorithms for real-time bayesian network inference. Laboratory for Knowledge Discovery in Databases Department of Computing and Information Sciences, Kansas State University, 2002, 20 р.

AttachmentSize
2013-2-6.pdf418.83 KB

Тематичні розділи журналу

,