Constructing financial distress prediction model using group method of data handling technique

Authors:

Abstract
Companies in financial distress make the creditors, shareholders, employees, investors and other participants of the related firms suffer great losses. In order to prevent the companies run into bankruptcy, financial distress prediction has been a useful tool for distinguishing companies in financial distress from those healthy. Statistical methods and artificial intelligence techniques have been widely used to deal with this issue. Many studies indicated that artificial neural networks outperform many statistical methods. However, artificial neural networks have the drawback of failing to interpret the classification results. This paper uses an artificial intelligence technique-group method of data handling technique to overcome this drawback. The sample data are collected from Taiwan listed companies in the Taiwan Stock Exchange Corporation. The result illustrates that the accuracy rates of classification of group method of data handling models are larger than 90% and the models of the group method of data handling obtain better accuracy than the models of discriminant analysis and logistic regression.

This publication has 27 references indexed in Scilit: