Comparing the Performance of Deep Learning Methods to Predict Companies’ Financial Failure

Abstract

One of the most crucial problems in the field of business is financial forecasting. Many companies are interested in forecasting their incoming financial status in order to adapt to the current financial and business environment to avoid bankruptcy. In this work, due to the effectiveness of Deep Learning methods with respect to classification tasks, we compare the performance of three well-known Deep Learning methods (Long-Short Term Memory, Deep Belief Network and Multilayer Perceptron model of 6 layers) with three bagging ensemble classifiers (Random Forest, Support Vector Machine and K-Nearest Neighbor) and two boosting ensemble classifiers (Adaptive Boosting and Extreme Gradient Boosting) in companies' financial failure prediction. Because of the inherent nature of the problem addressed, three extremely imbalanced datasets of Spanish, Taiwanese and Polish companies' data have been considered in this study. Thus, five oversampling balancing techniques, two hybrid balancing techniques (undersampling-oversampling) and one clustering-based balancing technique have been applied to avoid this data inconsistency problem. Considering the real financial data complexity level and type, the obtained results show that Multilayer Perceptron model of 6 layers, in conjunction with SMOTE-ENN balancing method, yielded the best performance according to the Accuracy, Recall and Type II error metrics. In addition, Long-Short Term Memory, Multilayer Perceptron and ensemble methods obtained also very good results, outperforming several classifiers used in previous studies with the same datasets.INDEX TERMS Economic forecasting, classification algorithms, machine learning, deep learning, data balancing.

Balancing Techniques for Advanced Financial Distress Detection Using Artificial Intelligence

“…SMOTE-ENN, SMOTE-TL, Spider, etc. [121,123,[133][134][135] The increase in training time and computational resources.…”

“…Numerous individual trees are created, which have a low correlation with one another. In addition, the majority of these trees' votes decide the class's label [123,203]. 8.…”

Bankruptcy prediction using ensemble of autoencoders optimized by genetic algorithm

“…The efficiency of autoencoders in terms of bankruptcy prediction was proven by Soui et al (2020) , where stacked autoencoders with the softmax classifier noticeably outperformed reference methods such as an SVM, a DT, LR, and an NN. Again, the authors pointed out the problem of interpretability, which is crucial in the field of bankruptcy prediction ( Aljawazneh et al, 2021 ) also compared various deep learning techniques for bankruptcy prediction tasks. The application of a multilayer perceptron with six layers combined with the SMOTHE_ENN balancing technique was determined to be the best method according to various metrics and its lowest misclassification rate.…”

State Regulator’s Role in the Country’s Banking System During Wartime

“…Besides, Li [4] examines the early prevention of the financial system crisis in the context of a number of indicators (innovation potential, ability to generate cash flows, profitability, operational efficiency, solvency, capital structure). In this regard, Aljawazneh et al [5] consider the tools of Deep Learning Methods in the context of analysing the performance of the corporate sector in terms of financial insolvency in view of exogenous and endogenous challenges. Theoretical and practical developments in this area are used in this study for selecting a set of key indicators for the analysis of the banking sector in conditions of significant turbulence during the period of armed aggression.…”