RISK ANALYSIS IN BAYESIAN NETWORKS: AN APPLICA-TION ON BANKING SECTOR
Year 2013,
Volume: 6 Issue: 1, 1 - 14, 14.08.2014
Emre Dünder
,
Mehmet Cengiz
,
Haydar Koç
Nurettin Savaş
Abstract
Bayesian Networks are the mathematical objects that represent uncertain relational structures in a multivariate dataset. Bayesian networks are widely used in many fields for being suitable in terms of inference. Especially in recent years Bayesian Networks are implemented such as in insurance, risk management and fraud detection. To construct a Bayesian network, structure and parameter learning algorithms are employed. In this study the performance of constructed Bayesian networks structures
are compared using a dataset on risk analysis. After constructing the Bayesian network structure the conditional probability values are calculated and according to these values, the assessments are made about credit risk position and the other variables. The applications are performed using GenIe and bnlearn package existing in
R programme.
References
- Fenton N. E., Neil M. D., 2012. Risk Assessment and Decision Analysis with Bayesian Networks, CRC Press.
- Gemela J., 2001. Financial analysis using Bayesian Networks, Applied Stochastıc Models in Busıness and Industry; 17:57-67.
- Neapolitan, R.E., Learning Bayesian Networks, 2003, Prentice Hall, Upper Saddle River, NJ.
- Margaritis, D., 2003. Learning Bayesian Network Model Structure from Data. PhD thesis, School of Computer Science, CarnegieMellon University, Pittsburgh, PA. CMU-CS 03-153.
- Oliva G. M., Weber P., Simon C., Iung B., 2009. Bayesian networks Applications on Dependability, Risk Analysis and Maintenance, 2nd IFAC Workshop on Dependable Control of Discrete System, DCDS'09, Bari : Italy .
- Scutari, M. 2011. Measures of Variability for Graphical Models. PhD thesis, Universita degli Studi di Padova, Dipartimento di Scienze Statistiche.
- Spirtes, P., Glymour, C., and Scheines, R., 2000. Causation, Prediction, and Search. IT Press.
- Tsamardinos, I., Aliferis, C., F., Statnikov, A., 2003. Algorithms for Large Scale Markov Blanket Discovery. In Proceedings of the 16th International Florida Artificial Intelligence Research Society Conference 376 381.
- Verma, T., S., Pearl, J. 1991. Equivalence and Synthesis of Causal Models.Uncertainty in Artifcial Intelligence, 6, 255-268.
- Yaramakala, S., Margaritis, D., 2005. Speculative Markov Blanket Discovery for Optimal Feature Selection, In Proceedings of the 5th IEEE International Conference on Data Mining ,pages 809-812.
BAYESCİ AĞLARDA RİSK ANALİZİ: BANKACILIK SEKTÖRÜ ÜZERİNE BİR UYGULAMA
Year 2013,
Volume: 6 Issue: 1, 1 - 14, 14.08.2014
Emre Dünder
,
Mehmet Cengiz
,
Haydar Koç
Nurettin Savaş
Abstract
Bayesci ağlar çok değişkenli bir veri seti içindeki belirsiz ilişkisel yapıları grafiksel modeller ile betimleyen matematiksel nesnelerdir. Bayesci ağlar çıkarsama açısın-dan elverişli olduğu için birçok farklı alanda yaygın biçimde kullanılmaktadır. Özel-likle son yıllarda sigortacılık, risk yönetimi ve suiistimal tespiti konularında Bayesci ağlar uygulanmaktadır. Bir Bayesci ağı oluşturmak için yapı öğrenme ve parametre öğrenme algoritmaları kullanılmaktadır. Bu çalışmada risk analizi üzerine bir veri seti kullanılarak oluşturulan Bayesci ağ yapılarının performansı karşılaştırılmıştır. Bayesci ağ yapısı oluşturulduktan sonra koşullu olasılık değerleri hesaplanmış ve bu değerlere göre kredi risk durumu ve diğer değişkenlere ilişkin değerlendirmeler yapılmıştır. Uygulamalar GeNIe programı ve R programı içerisinde bulunan bnlearn paketi kullanılarak gerçekleştirilmiştir.
References
- Fenton N. E., Neil M. D., 2012. Risk Assessment and Decision Analysis with Bayesian Networks, CRC Press.
- Gemela J., 2001. Financial analysis using Bayesian Networks, Applied Stochastıc Models in Busıness and Industry; 17:57-67.
- Neapolitan, R.E., Learning Bayesian Networks, 2003, Prentice Hall, Upper Saddle River, NJ.
- Margaritis, D., 2003. Learning Bayesian Network Model Structure from Data. PhD thesis, School of Computer Science, CarnegieMellon University, Pittsburgh, PA. CMU-CS 03-153.
- Oliva G. M., Weber P., Simon C., Iung B., 2009. Bayesian networks Applications on Dependability, Risk Analysis and Maintenance, 2nd IFAC Workshop on Dependable Control of Discrete System, DCDS'09, Bari : Italy .
- Scutari, M. 2011. Measures of Variability for Graphical Models. PhD thesis, Universita degli Studi di Padova, Dipartimento di Scienze Statistiche.
- Spirtes, P., Glymour, C., and Scheines, R., 2000. Causation, Prediction, and Search. IT Press.
- Tsamardinos, I., Aliferis, C., F., Statnikov, A., 2003. Algorithms for Large Scale Markov Blanket Discovery. In Proceedings of the 16th International Florida Artificial Intelligence Research Society Conference 376 381.
- Verma, T., S., Pearl, J. 1991. Equivalence and Synthesis of Causal Models.Uncertainty in Artifcial Intelligence, 6, 255-268.
- Yaramakala, S., Margaritis, D., 2005. Speculative Markov Blanket Discovery for Optimal Feature Selection, In Proceedings of the 5th IEEE International Conference on Data Mining ,pages 809-812.