Sieci neuronowe i polichotomiczne modele zmiennych jakościowych w analizie ryzyka kredytowego

• Autorzy: Paweł Baster , Katarzyna Pocztowska

Warianty tytułu

Neural Networks and Models for Polychotomous Ordered Data in Credit Risk Analysis

• Języki publikacji: PL

Abstrakty

EN

Management of credit risk, one of the main bank activities, is currently a very important issue. This paper contains comparison of two instruments used in prediction of probability that consumer fails to fully repay a loan in agreed time: artificial neural networks and models for polychotomous ordered data. For the empirical research each client has been assigned to one of four categories reflecting his/her delay in payments. Estimation and validation of methods was performed on a 3000-item sample containing information about each loan agreement and repayment history originating from one of Polish banks, covering years 2000-2001. The dataset was repeatedly divided into train and validation sets. Multi-layer architecture of artificial neural network with logistic activation function was proposed. Ordered logit and probit models were estimated within maximum likelihood framework. Several alternative specifications were proposed differing in independent variable set (including their products and squares). Bank income was chosen as the main criterion of fitness. Problem of optimal decision and defining appropriate loss function was formulated on the basis of statistical decision theory. Furthermore, properties of estimated models related to inference about probability of repayment and credit risk factors were presented. (original abstract)

Słowa kluczowe

PL

Sieci neuronowe; Zmienne jakościowe; Analiza ryzyka kredytowego

EN

Neural networks; Qualitative variables; Credit risk analysis

• Czasopismo: Folia Oeconomica Cracoviensia
• Rocznik: 2011
• Tom: 52
• Strony: 5--17

Twórcy

autor

Paweł Baster

autor

Katarzyna Pocztowska

Bibliografia

• G. Cybenko, Approximations by superpositions of sigmoidal function, Mathematics of Control, Signals and Systems 2, 1989, 303-314
• M. Gruszczynski, Modele i prognozy zmiennych jakościowych w finansach i bankowości, Monografie i Opracowania SGH 6, Warszawa 2001
• Z. Krysiak, Ryzyko kredytowe a wartość firmy. Pomiar i modelowanie, Oficyna Wydawnicza, Kraków 2006
• H. Lee, Model Selection and Model Averaging for Neural Networks, PhD thesis, Carnegie Mellon University, Pittsburgh 1999, USA. http://users.soe.ucsc.edu/~herbie/thesis.pdf
• J. Marzec, Bayesowskie modele zmiennych jakościowych i ograniczonych w badaniach niespłacalności kredytów, Wydawnictwo UE w Krakowie, Kraków 2008
• J. Osiewalski, Bayesowska statystyka i teoria decyzji w analizie ryzyka kredytu detalicznego [w:] Finansowe warunkowania decyzji ekonomicznych, Krakowska Szkoła Wyższa im. Andrzeja Frycza Modrzewskiego, Kraków 2007
• B. Petrozolin-Skowrońska, Neuronowe sieci. [w:] Nowa Encyklopedia Powszechna PWN, t. 4, Wydawnictwo Naukowe PWN, Warszawa 1996, 439
• R.D. McKelvey, W. Zavoina, A Statistical Model for the Analysis of Ordinary Level Dependent Variables, Journal of Mathematical Sociology 4, 1975, 103-120
• R. Tadeusiewicz, Sieci Neuronowe, Akademicka Oficyna Wydawnicza, Kraków 1993