Financial bankruptcy prediction: An application in the BIST

In the first chapter, after giving information about the corporate goal and its theories, the importance of corporate management, financial management, financial planning and financial analysis and performance evaluation and their importance for corporate organizations are mentioned. In the second part, the financial failure definition, causes, consequences, effects, ways to prevent, the crisis experienced in the world and their impact on Turkey are explained. Then the importance of financial failure prediction is given and related literature search results and financial failure prediction studies in the world and Turkey are presented. In the third chapter, the traditional and modern prediction methods used in the financial failure studies are explained and the reasons and the superior aspects of the modern techniques are explained. In the fourth chapter, financial failure prediction has been performed to compare predictability performance of traditional and modern prediction methods. For this purpose, the financial failures of manufacturing companies which were traded between 1997-2018 in Borsa Istanbul were tried to be predicted 3 months in advance. For this aim, logistic regression was performed as a traditional method and artificial neural network model, decision tree and random forest were applied as modern prediction method. As a result of the study, it has been concluded that modern methods perform better than traditional methods in predicting financial failure. The most successful prediction models were Artificial Neural Network (ANN), Random Forests (RF), Decision Trees (DT) and Logistic Regression (LR) model. As a result of the study, it has been concluded that modern methods perform better than traditional methods in predicting financial failure. The most successful prediction models were the modern Artificial Neural Network (ANN), Random Forests (RF), Decision Trees (DT), respectively. The most unsuccessful model was the traditional Logistic Regression (LR) model. At the end of the study, suggestions regarding the difficulties we encountered during the implementation and proposals for future financial failure prediction are presented.