A MLP based PD estimation model for SME credits
2009 24th International Symposium on Computer and Information Sciences, ISCIS 2009, Guzelyurt, Kıbrıs (Gkry), 14 - 16 Eylül 2009, ss.265-268, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/iscis.2009.5291824
- Basıldığı Şehir: Guzelyurt
- Basıldığı Ülke: Kıbrıs (Gkry)
- Sayfa Sayıları: ss.265-268
- Anahtar Kelimeler: Credit risk analysis (CRA), Logistic regression (LR), Multilayer perceptron (MLP), Probability of default (PD), SME
- Boğaziçi Üniversitesi Adresli: Evet
Özet
This study presents a multilayer perceptron (MLP) cascaded with a logistic regression (LR) model for the probability of default (PD) estimation of real-life Small and Medium Enterprises (SMEs). The MLP does a preprocessing for LR which is used to compute PD value. The obtained raw PD values are calibrated according to the real portfolio default average. In experiments, a Turkish SME database is used. The results indicate that the cascaded MLP-LR model provides better classification accuracy and outperforms commonly used logistic regression. © 2009 IEEE.