Comparison of financial distress prediction models: Evidence from Turkey

Abstract

The purpose of this paper is to explore the differences and similarities between financial distress prediction (FDP) models and to determine which explanatory variables and methodologies are the most effective in prediction of financial distress. For this purpose, 167 manufacturing companies (full sample) listed in Istanbul Stock Exchange (ISE) were used. In total, 27 financial ratios were identified from previous literature studies as potentially significant and they were calculated for the years 2009 and 2010. In the study, logistic regression, artificial neural networks and decision tree methods, which are frequently used in the literature, have been employed. As a result, many of the financial ratios are found to be effective in predicting financial distress. Moreover, logistic regression and artificial neural network methods have indicated better prediction accuracy results of financial distress for classification of companies. © EuroJournals Publishing, Inc. 2012.