Corporate default prediction with adaboost and bagging classifiers

Abstract

This study aims to show a substitute technique to corporate default prediction. Data mining techniques have been extensively applied for this task, due to its ability to notice non-linear relationships and show a good performance in presence of noisy information, as it usually happens in corporate default prediction problems. In spite of several progressive methods that have widely been proposed, this area of research is not out dated and still needs further examination. In this paper, the performance of ensemble classifier systems is assessed in terms of their capability to appropriately classify default and non-default Malaysian firms listed in Bursa Malaysia. AdaBoost and Bagging are novel ensemble learning algorithms that construct the base classifiers in sequence using different versions of the training data set. In this paper, we compare the prediction accuracy of both techniques and single classifiers on a set of Malaysian firms, considering the usual predicting variables such as financial ratios. We show that our approach decreases the generalization error by about thirty percent with respect to the error produced with a single classifier.