Levent GÜNTAY, Mehmet AKTUNA

Finansal Kurumlarda Senaryo Bazlı Aykırı Gözlem Tespiti: Türk Faktoring Sektörü Üzerine Bir Çalışma

Finans sektöründe çevrimiçi ve mobil işlemlerin sayısı ve hızının artması beraberinde farklı riskleri ve denet- leme maliyetlerini de getirmiştir. Bu riskler sahtecilikten kredi riskine, veri tabanı hatalarından, operasyonel problemler ve müşteri kayıplarına kadar çok farklı alanlarda gerçekleşebilir. Bu çalışmada faktoring işlemleri için senaryo bazlı aykırılık analizi bu riskleri oluşma aşamasında ve gözetimli bir istatistiksel bir model kurma- dan tespit etmeyi amaçlamaktadır. Aykırılık analizi bağlamında karakteristikleri ana kümeden büyük sapma gösteren çek, müşteri ya da müşteri temsilcisi gözlemleri aykırı olarak tanımlanmaktadır. Bu karakteristikler fak- toring uzmanlarının tecrübelerine dayanılarak geliştirilen senaryo kurguları içinde seçilip bir araya getirilmiştir. Karakteristiklerin ana kümeden sapmaları Mahalanobis, Minimum Kovaryans, ve Ortogonolize Gnanadesikan- Kettenring uzaklıkları ile hesaplanmaktadır. Çalışmada kullanılan veritabanı bir faktoring şirketinin 2018-2020 arası çek faktoring işlem bilgileri ile Kredi Kayıt Bürosu Çek ve Risk raporlarını birleştirmekte ve 7 farklı senaryo kullanılarak aykırı işlemler bulunmaktadır. Kurulan modelin aykırı değer eşik seviyesinin finansal kurumun to- lere edebileceği hata tespit oranları ve istihbarat bütçesi çerçevesinde nasıl ayarlanıp optimize edilebileceği de çalışmada gösterilmiştir. Geliştirilen model bankacılık, faktoring, leasing, sigortacılık alanlarındaki hemen her finansal işlemde risk taşıyan aykırı gözlemleri bulabildiği gibi finansal sektörü düzenleyici ve denetleyici kurumlar tarafından da kullanılabilir.

Scenario Based Anomaly Detection in Financial Institutions: A Study on the Turkish Factoring Sector

The increase in the number and speed of online and mobile transactions in the financial sector generates various risks and monitoring costs. Some of these risks include fraud risk, credit risk, database errors, operation- al problems and churn risk. In this study, scenario-based anomaly detection analysis for factoring transactions is used to identify these risks at an early stage without establishing a supervised statistical model. In anomaly detection, observations at the check, customer or customer representative level whose characteristics deviate from the main cluster are defined as outliers. The characteristics in scenarios are selected based on the experi- ence of factoring experts. The deviations of the characteristics from the main cluster are calculated by the Ma- halanobis, Minimum Covariance, and Orthogonolized Gnanadesikan-Kettenring distances. The data used in this study are comprised of check-level factoring transactions of a factoring company between 2018-2020 and the check and risk reports issued by the Credit Registration Bureau and are detected as outliers by using 7 different risk scenarios. The study also shows that the outlier detection threshold can be optimized within the framework by considering the model errors and the monitoring budget of the financial institution. The developed model can detect risk carrying anomalies in almost every financial transaction in the banking, factoring, leasing, and insurance sectors and can be also employed by the financial regulatory and supervisory institutions

PDF

___

1. Alpu, Ö. (2016). Aykırı değer varlığında hızlı minimum kovaryans determi- nantı kestiricilerinin faktör analizinde kullanımı. Sakarya University Journal of Science, 20(3), 701-709.
2. Ané, T., Ureche-Rangau, L., Gambet, J. B., & Bouverot, J. (2008). Robust outlier detection for Asia–Pacific stock index returns. Journal of International Financial Markets, Institutions and Money, 18(4), 326-343.
3. Aydın, O. M., & Aktaş, R. (2020) Detecting Financial Information Manipula- tion By Using Supervised Machine Learning Technics: SVM, PNN, KNN, DT. Uluslararası İktisadi ve İdari İncelemeler Dergisi, (29), 165-174.
4. Breunig, M. M., Kriegel, H. P., Ng, R. T., & Sander, J. (2000). LOF: identifying density-based local outliers. In Proceedings of the 2000 ACM SIGMOD inter- national conference on Management of data (pp. 93-104).
5. Campello, R. J., Moulavi, D., Zimek, A., & Sander, J. (2015). Hierarchical den- sity estimates for data clustering, visualization, and outlier detection. ACM Transactions on Knowledge Discovery from Data (TKDD), 10(1), 1-51.
6. Cao, B., Mao, M., Viidu, S., & Yu, P. (2018). Collective fraud detection cap- turing inter-transaction dependency. In Proceedings of the KDD 2017 Work- shop on Anomaly Detection in Finance (pp. 66-75).
7. Chawla, S., & Gionis, A. (2013). k-means–: A unified approach to clustering and outlier detection. In Proceedings of the 2013 SIAM International Con- ference on Data Mining (pp. 189-197). Society for Industrial and Applied Mathematics.
8. Cho, S., Hong, H., & Ha, B. C. (2010). A hybrid approach based on the com- bination of variable selection using decision trees and case-based reasoning using the Mahalanobis distance: For bankruptcy prediction. Expert Systems with Applications, 37(4), 3482-3488.
9. Çetiner, M., Dinçsoy, Ö., & Toraman, T. (2020, October). Outlier Detection for Analysis of Real Estate Price. In 2020 28th Signal Processing and Commu- nications Applications Conference (SIU) (pp. 1-4). IEEE.
10. Devlin, S. J., Gnanadesikan, R., & Kettenring, J. R. (1975). Robust estimation and outlier detection with correlation coefficients. Biometrika, 62(3), 531- 545.
11. Devlin, S. J., Gnanadesikan, R., & Kettenring, J. R. (1981). Robust estimation of dispersion matrices and principal components. Journal of the American Statistical Association, 76(374), 354-362.
12. Duan, L., Xu, L., Liu, Y., & Lee, J. (2009). Cluster-based outlier detection. An- nals of Operations Research, 168(1), 151-168.
13. Esen, M. F., & Timor, M. (2019). Çok Değişkenli Aykırı Değer Tespiti İçin Kla- sik Ve Dayanıklı Mahalanobis Uzaklık Ölçütleri: Finansal Veri İle Bir Uygulama. Uluslararası İktisadi ve İdari İncelemeler Dergisi, (25), 267-282.
14. Eskin, E., Arnold, A., Prerau, M., Portnoy, L., & Stolfo, S. (2002). A geometric framework for unsupervised anomaly detection. In Applications of data mi- ning in computer security (pp. 77-101). Springer, Boston, MA.
15. Gnanadesikan, R., & Kettenring, J. R. (1972). Robust estimates, residuals, and outlier detection with multiresponse data. Biometrics, 81-124.
16. Jiang, S. Y., & An, Q. B. (2008). Clustering-based outlier detection method. In 2008 Fifth International Conference on Fuzzy Systems and Knowledge Discovery (Vol. 2, pp. 429-433). IEEE.
17. Karminsky, A. M., & Khromova, E. (2018). Increase of banks’ credit risks fore- casting power by the usage of the set of alternative models. Russian Journal of Economics, 4, 155.
18. Knox, E. M., & Ng, R. T. (1998). Algorithms for mining distancebased outliers in large datasets. In Proceedings of the international conference on very lar- ge data bases (pp. 392-403).
19. Kriegel, H. P., Schubert, M., & Zimek, A. (2008). Angle-based outlier detecti- on in high-dimensional data. In Proceedings of the 14th ACM SIGKDD inter- national conference on Knowledge discovery and data mining (pp. 444-452).
20. Mahalanobis, P. C. (1936). On the generalised distance in statistics. Procee- dings of the National Institute of Sciences of India, 1936, 49-55.
21. Maronna, R. A. (1976). Robust M-estimators of multivariate location and scatter. The annals of statistics, 51-67
22. Maronna, R. A., Martin, R. D., Yohai, V. J., & Salibián-Barrera, M. (2019). Ro- bust statistics: theory and methods (with R). John Wiley & Sons.
23. Maronna, R. A., & Zamar, R. H. (2002). Robust estimates of location and dispersion for high-dimensional datasets. Technometrics, 44(4), 307-317.
24. Moh’d Belal, A. Z., Al-Dahoud, A., & Yahya, A. A. (2010). New outlier detec- tion method based on fuzzy clustering. WSEAS transactions on information science and applications, 7, 681-690.
25. Prakash, A., & Chandrasekar, C. (2012). A novel hidden Markov model for credit card fraud detection. International Journal of Computer Applicati- ons, 59(3), p35-41.
26. Rahmani, A., Afra, S., Zarour, O., Addam, O., Koochakzadeh, N., Kianmehr, K., & Rokne, J. (2014). Graph-based approach for outlier detection in sequen- tial data and its application on stock market and weather data. Knowledge- Based Systems, 61, 89-97.
27. Ramaswamy, S., Rastogi, R., & Shim, K. (2000). Efficient algorithms for mi- ning outliers from large data sets. In Proceedings of the 2000 ACM SIGMOD international conference on Management of data (pp. 427-438).
28. Rousseeuw, P. J. (1985). Multivariate estimation with high breakdown po- int. Mathematical statistics and applications, 8 (283-297), 37.
29. Rousseeuw, P. J., & Van Zomeren, B. C. (1990). Unmasking multivariate outliers and leverage points. Journal of the American Statistical associati- on, 85(411), 633-639.
30. Rousseeuw, P. J., & Driessen, K. V. (1999). A fast algorithm for the minimum covariance determinant estimator. Technometrics, 41(3), 212-223.
31. Rousseeuw, P. J., & Leroy, A. M. (2005). Robust regression and outlier detec- tion (Vol. 589). John wiley & sons.
32. Sánchez, D., Vila, M. A., Cerda, L., & Serrano, J. M. (2009). Association ru- les applied to credit card fraud detection. Expert systems with Applications, 36(2), 3630-3640.
33. Tang, B., & He, H. (2017). A local density-based approach for outlier detecti- on. Neurocomputing, 241, 171-180.
34. Weekley, R. A., Goodrich, R. K., & Cornman, L. B. (2010). An algorithm for classification and outlier detection of time-series data. Journal of Atmosphe- ric and Oceanic Technology, 27(1), 94-107.
35. Wiens, D. P., & Zheng, Z. (1986). Robust M‐estimators of multivariate loca- tion and scatter in the presence of asymmetry. Canadian Journal of Statis- tics, 14(2), 161-176.