Prediction of Financial Distress for Tunisian Firms: A Comparative Study Between Financial Analysis and Neuronal Analysis

This paper presents a prognosis of financial distress of Tunisian firms. For the purpose, we empirically compared the financial analysis to artificial neural network analysis. Five multilayer perceptron are applied to improve banking-decision. Based on the results of correct classification rate, artificial neural network proved an intact predictive ability. As well, the findings of generalization test confirmed the conclusion of the classical financial analysis of a company not included in our base sample. The artificial neural network can effectively automate the granting credit decision then performed better than traditional financial analysis.