Predicting Bankruptcy with Support Vector Machines

Fuzzy Support Vector Machine for bankruptcy prediction

“…SVM is based on the idea of SRM [36,63,71] to build a model of a given system. The foundations of SVM have been developed by Vapnik [71].…”

“…An effective prediction in time is valued priceless for business in order to evaluate risks or prevent bankruptcy [4,5,20]. A fair amount of research has therefore focused on bankruptcy prediction [1,2,6,7,9,10,[13][14][15][16]18,23,25,30,33,[35][36][37][40][41][42][43]45,50,52,53,55,56,58,59,61,63,64,67,69,70,74,81,83]. diction models to deter disastrous consequences of ultimate financial distress.…”

“…The third group encompasses different Computational Intelligence [42] techniques such as decision trees, ANN [38], Support Vector Machines (SVM) [36,63] etc. Most researchers use one of techniques to compare prediction performance with other techniques for specific data set [5,17,[36][37][38][39]43,63,72]. However, there is no single conclusion that one technique is consistently better than another for general bankruptcy prediction.…”

“…SVM has been studied as bankruptcy prediction tool quite often in past [36,38,42,48,55,56,63]. It is a classification method based on Statistical Learning (SL) theory.…”

Credit risk prediction using support vector machines

“…The resulting classifier generalizes well even in high dimensional input spaces and under small training sample conditions. As the amount of available financial default data is typically small and often of low quality, this classifier seems particularly suitable for credit risk estimation, including default classification (Härdle et al 2005) and PD estimation. This paper gives a comprehensive introduction to SVM and compares the credit rating capabilities of a SVM classifier to a LRM regarding default classification and PD prediction.…”