Statistical and Machine Learning Approaches in Credit Scoring
This paper and another related to it (Jafar-Shaghaghi (1992)) are intended to compare two classification methods, namely discriminant analysis in statistics with classification in machine learning through their application in the area of economics. This paper uses credit assessment as an area for applying and analysing these two methods. In section two the theoretical background of this topic will be described. Section 3 is devoted to credit assessment methods. Empirical and theoretical evaluation will be discussed in section 4.
Unable to display preview. Download preview PDF.
- Bertzger, T.M. (1991), “Die Anwendung statistischer Verfahren zur Risikofrüherkennung bei Dispositionskredit” Studienreihe der Stiftung Kreditwirtschaft an der Universität Hohenheim, Bd. 9.Google Scholar
- Carter, C. and Catlett, J. (1987), “Assessing Credit Card Application Using Hachine Learning”, in: IEEE Expert, Fall 1987, 71–79.Google Scholar
- Jafar-Shaghaghi, F. (1992), “Theoretical and Empirical Evaluation of some Inductive Based Hethods in Artificial Intelligence and Statistics”,in Operations Research 91, Gritzmann, P. et al (eds) Phisica-Verlag, Heidelberg, pp 519–522.Google Scholar
- Lachenbruch, P.A. (1975), “Disriminant Analysis”, Hafner, New York.Google Scholar
- Quinlan, J.R. (1986), “Induction of Decision Trees”, Machine Learning, Vol. 1, Nr. 1.Google Scholar
- Qiunlan, J.R. (1987), “Decision Trees and Hultivariate Attributes”, in Machine Intelligence 11, J.E. Hays and D. Michie (eds), Oxford University Press.Google Scholar
- Assistant Profesional is a machine learning tool (developed at the Josef Stefan Institute of Ljubljana university).Google Scholar
- IXL is a machine learning tool (IntelligenceWare, Inc. 5933 West Century Elv. Los Angeles, CA, 90045).Google Scholar