Financial failure prediction of companies using artificial i̇ntelligence methods: An application in Bist manufacturing sector

The financial failure affects negatively on a country's economical growth with the increase in the number of businesses as it threats their future. It is significantly important to foresee the failure so to take precautions as its result and get out of the problems for the businesses. Many models were developed about the estimation of financial failure. These models are mostly statistical and artificial intelligence techniques. The estimations of financial failure were made with the use of artificial neural networks, support vector machines and ensemble learning models in this study. 140 companies which are dealt/ were dealt in the manufacturing sector in Istanbul Stock Exchange were received as the sample. 26 financial rates which are frequently used in the literature were used as the model's input variables. The model's classification performances were compared in the study, and the accuracy, specifity and sensitivity percentages were calculated for that classification. Moreover, the significance values of model related to 26 financial rates which constitute the study's variables were calculated. The performances of estimation models were measured with ROC curves which were used in the classification problems. As a result of the study, while the artificial neural networks had a better classification performance than the support vector machines and the ensemble had a better classification than the other two machine learning models.