TÜRK BANKACILIK SEKTÖRÜNDE FİNANSAL GÜÇ DERECELERİNİN TAHMİNİNDE YAPAY SİNİR AĞLARI VE ÇOK DEĞİŞKENLİ İSTATİSTİKSEL ANALİZ TEKNİKLERİNİN PERFORMANSLARININ KARŞILAŞTIRILMASI

Öz

Finansal güç derecelendirmesi (financial strength rating), bir bankanın temel finansal gücünü gösterir. Burada amaçlanan bir bankanın temel finansal gücünün dış faktörlerin değerlendirme harici bırakılması suretiyle ölçülmesidir. Dış faktörler, bankanın faaliyet çevresinden kaynaklanabileceği gibi, koruyucu nitelikteki dış destek mekanizmalarının varlığı ile de bağlantılı olabilir. Yapılan değerlendirme ile bankanın, koruyucu dış faktörlerden tamamen arındırılmış derecelendirmesi nasıl olurdu sorusuna cevap aranır. Ayrıca bu değerlendirmede bankanın finansal temeli, şube ağının gücü, faaliyet alanlarındaki ve varlıklarındaki çeşitlilik incelenir.Bu çalışmada Türk bankalarının finansal güç derecelerini yapay sinir ağları ve çok değişkenli istatistiksek analiz teknikleri kullanarak tahmin etmek amacıyla bir model geliştirilmiştir. Çalışmanın metodolojisi, modelde yer alan değişkenlerin seçilmesi, veri setinin oluşturulması, kullanılacak tekniklerin belirlenmesi ve bu tekniklerin doğru sınıflandırma başarısının değerlendirilmesinden oluşmaktadır. Yapay sinir ağı, modelin elde edildiği veri setinde çok değişkenli istatistiksek analiz teknikleri ne göre yüksek bir sınıflandırma performansı göstermiştir. Modelin geçerliliğinin test edildiği veri setinde ise kullanılan tekniklerin tahmin performansları arasında anlamlı bir fark bulunamamıştır.

Anahtar Kelimeler

• Bankalar, Finansal Güç Derecelendirmesi, Yapay Sinir Ağları, Çok Değişkenli İstatistiksel Analiz Teknikleri

A PERFORMANCE COMPARISON BETWEEN ARTIFICIAL NEURAL NETWORKS AND MULTIVARIATE STATISTICAL METHODS IN FORECASTING FINANCIAL STRENGTH RATING IN TURKISH BANKING SECTOR

Öz

Financial strength rating indicates the fundamental financial strength of a bank. The aim of financial strength rating is to measure a bank’s fundamental financial strength excluding the external factors. External factors can stem from the working environment or can be linked with the outside protective support mechanisms. With the evaluation, the rating of a bank free from outside supportive factors is being sought. Also the financial fundamental, franchise value, the variety of assets and working environment of a bank are being evaluated in this context.In this study, a model has been developed in order to predict the financial strength rating of Turkish banks. The methodology of this study is as follows: Selecting variables to be used in the model, creating a data set, choosing the techniques to be used and the evaluation of classification success of the techniques. It is concluded that the artificial neural network system shows a better performance in terms of classification of financial strength rating in comparison to multivariate statistical methods in the raining set. On the other hand, there is no meaningful difference could be found in the validation set in which the prediction performances of the employed techniques are tested.

Anahtar Kelimeler

• Banks, Financial Strength Rating,,Artificial Neural Networks, Multivariate Statistical Methods

Kaynakça

1. ALTMAN, E. I., S. KATZ (1976), “An Analysis of Bond Ratings in the Electric Public Utility Industry”, Proceeding of the Conference on Topical Research in Accounting, G. Sorter, M. Schiff (Ed.), Ross Institute, New York University.
2. ANG, J. S., P. A. KIRITKUMAR (1978), “Bond Rating Methods: Comparison and Validation”, The Journal of Finance.
3. BAETGE, J. (1994), “Rating von Unternehmen anhand von Bilanzen”, Die Wirtschaftsprüfung, Heft 1.
4. BAETGE, J., A. JERSCHENSKY (1996), “Beurteilung der wirtschaftlichen Lage von Unternehmen mit Hilfe von modernen Verfahren der Jahresabschlussanalyse”, Der Betrieb, Heft 32.
5. BELKAOUI, A. (1980), “Industrial Bond Ratings, A New Look”, Journal of Financial Management, Vol. 9, No.3.
6. BELKAOUI, A. (1983), “Industrial Bonds and The Rating Process”, CT: Quorum Books, Wesport.
7. BURGER, A., B. SCHELLBERG (1994), “Rating von Unternehmen mit neuronalen Netzen”, Betriebs-Berater, Heft 13.
8. BURGER, A., A. BUCHHART (1998), “Rating und Risikokosten im Kreditgeschaeft”, Die Bank, Heft 7.

9. CHAVEESUK, R., C. SRIVAREE-RATANA, A. E. SMITH (1999), “Alternative Neural Network Approaches to Corporate Bond Rating”, Journal of Engineering Valuation and Cost Analysis , Vol. 2, No. 2.
10. CHU, C.H. (1997), “An Improved Neural Network for Manufacturing Cell Formation”, Decision Support Systems, Vol. 20.
11. CRYSTAL, J. S. B., G. DAGES, L. S. GOLDBERG (2001), “Does Foreign Ownership Contribute to Sounder Banks in Emerging Markets? The Latin American Experience”, Federal Reserve Bank of New York.
12. DUTTA, S., S. SHEKHAR (1988), “Bond Rating: A Non Conservative Application of Neural Networks”, Proceedings of the IEEE International Conference on Neural Netwoks, Vol. 2, San Diego.
13. EDERINGTON, L. H. (1986), “Why Split Ratings Occur”, Financial Management, Vol. 15.
14. FISHER, F. (1959), “Determinants of Risk Premiums on Corporate Bond”, Journal of Political Economy, June.
15. FONS, J. S. (1998), “Improving Transparency in Asian Banking Systems”, Moody's Investors Service.
16. GENTRY, J., P. NEWBOLD, D. WHITFORD (1988), “Predicting Industrial Bond Ratings with a Probit Model and Funds Flow Components”, Financial Review, Vol. 23, No. 3.
17. HORRIGAN, J. O. (1966), “The Determination of Long-Term Credit Standing with Financial Ratios”, Empirical Research in Accounting: Selected Studies, Supplement to Journal of Accounting Research, Vol. 4.
18. HUANG, Z., H. CHEN, C.J. HSU, W.H. CHEN,, S. W. (2004), “Credit Rating Analysis with Support Vector Machines and Neural Networks: A Market Comparative Study”, Decision Support Systems, Vol. 37.
19. KAPLAN, R., S. G. URWITZ (1979), “Statistical Models of Bond Ratings: A Methodological Inquiry, Journal of Business, Vol. 52, No. 2.
20. KIM, J. W. (1992), A Comparative Analysis of Rule-based, Neural Network and Statistical Classification Systems for The Bond Rating Problem, PhD, Virginia Commonwealth University, Richmond.
21. KIM, J. W., R. H. WEISTROFFER, R. T. REDMOND (1993), “Expert Systems for Bond Rating: A Comparative Analysis of Statistical, Rule-based and Neural Network Systems ”, Expert Systems, Vol. 10, No. 3.
22. MAHER, J. J., T. K. SEN (1997), “Predicting Bond Ratings Using Neural Networks: A Comparison with Logistic Regression ”, Intelligent Systems in Accounting, Finance and Management, Vol. 6.
23. MARTIN, L. J., G. V. HENDERSON (1983), “On Bond Ratings and Pension Obligations: A Note”, Journal of Financial and Quantitative Analysis, Vol. 18.
24. MOODY, J., J. UTANS (1995), “Architecture Selecgtion Strategies for Neural Networks Application to Corporate Bond Rating”, in: A. Refenes (Ed.), Neural Networks in the Capital Markets, Wiley, Chichester.
25. PERRY, L. G., HENDERSON G. V. , T. P. CRONAN (1984), “Multivariate Analysis of Corporate Bond Ratings and Industry Classification”, The Journal of Financial Research, Vol. 7, No. 1.
26. PINCHES, G. E., K. A. MINGO (1973), “A Multivariate Analysis of Industrial Bond Ratings”, Journal of Finance, Vol. 28, No. 1.
27. PINCHES, G. E., K. A. MINGO (1975), “The Role of Subordination and Industrial Bond Ratings”, Journal of Finance, Vol. 30, No. 1.
28. POGUE, T. F., R. M. SOLDOFSKY (1969), “What’s in A Bond Rating”, Journal of Financial and Quantitative Analysis, Vol. 4, No. 2.
29. POON, W. P. H., M. FIRTH, H. G. FUNG (1999), “A Multivariate Analysis of the Determinants of Moody’s Bank Financial Strength Ratings”, Journal of International Financial Markets, Institutions and Money, Vol. 9, No. 3.
30. RUMELHART D.E., J.L. MCCLELLAND, ve the PDP Group (1989), Parallel Distributed Processing: Exploration in the Microstructure of Cognition, Vol. 1, MIT Press, MA.
31. SINGLETON, J. C., A. J. S. SURKAN, (1990), “Neural Network for Bond Rating Improved by Multiple Hidden Layers”, Proceedings of the IEEE International Conference on Neural Netwoks, Vol. 2, San Diego.
32. TAN, R.P.G.H., J. VAN DEN BERG, W. M. VAN DEN BERGH (2002), “Credit Rating Classification Using Self Organizing Maps”, In: Smith, K., Gupta, J. (Eds.), Neural Networks in Business: Techniques and Applications, IDEA Group Publishing.
33. TBB (Mayıs 2002-2006), Bankalarımız 2001, 2002, 2003, 2004, 2005, Istanbul.
34. WEST, R. R. (1970), “An Alternative Approach to Predicting Corporate Bond Ratings”, Journal of Accounting Research, Vol. 8, No.1.
35. WEST, D. (2000), “Neural Network Credit Scoring Models”, Computers & Operations Research, Vol. 27.