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

H Aljawazneh,A M Mora,P A Castillo-Valdivieso,P Garcia-Sanchez

doi:10.1109/access.2021.3093461

H Aljawazneh, A M Mora + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3093461

Copy DOI

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 (oversampling-undersampling) and one clustering-based balancing technique have been applied to avoid data inconsistency problem. Considering the real financial data complexity level and type, the results show that the 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 and ensemble methods obtained also very good results, outperforming several classifiers used in previous studies with the same datasets.

Highlights

T HE problem of bankruptcy prediction has attracted the attention of researchers since the Crack of 1929 [1]
We present a novel comparison between three different Deep Learning (DL) methods, i.e., Deep Belief Network (DBN), Long-Short Term Memory (LSTM) and MULTILAYER PERCEPTRON WITH 6 LAYERS (MLP-6L), and five ensemble classifiers, i.e., Random Forest (RF), Support Vector Machine (SVM), KNearest Neighbor (KNN), AdaBoost and XGBoost, in predicting companies’ financial failure
We present a visual representation of all metrics values obtained by each combination of classifier and balancing technique addressed in this study, and applied to the Spanish, Taiwanese and Polish companies’ datasets in the Figures 14, 15 and 16, respectively, where it can be seen the high values reached by almost all the methods in most of the metrics

Summary

Introduction

T HE problem of bankruptcy prediction has attracted the attention of researchers since the Crack of 1929 [1]. The effects of bankruptcy on a company are of great significance, as they affect a large number of stakeholders, including workers, creditors and suppliers, and eventually, even entire countries. Machine Learning (ML), and more recently Deep Learning (DL) [2], have gained the interest of researchers in the financial area. The data related to the companies’ financial status are inherently imbalanced, since the bankruptcy is relatively uncommon in real life [3]. Several works have been focused on addressing the lack of patterns of minority classes, such as bankrupt companies in our problem, because it is dramatically affecting the classifiers, causing a decrease in their reliability and performance. We have considered the most appropriate and relevant and applied them to financial data

Objectives

Methods

Results

Conclusion