Bankruptcy Prediction Research Articles

A Machine Learning Based Framework For Bankruptcy Prediction In Corporate Finances Using Explainable AI Techniques

A Machine Learning Based Framework For Bankruptcy Prediction In Corporate Finances Using Explainable AI Techniques

Predicting Risk of Financial Distress using Grover’s Model: A Case study in Malaysia Companies

A bankruptcy prediction is one of the main critical problems for financial decision-makers. In this study, we aim to measure the risk of 5 Malaysian companies' financial failures through Grover's model and further investigate the accuracy of this model in predicting bankruptcy risk. Purposive sampling was used in this study, with five firms being chosen to be sampled in predicting bankruptcy risk. Meanwhile, the data from 292 US companies is used to test the accuracy of Grover's model in predicting bankruptcy risk. The predicted results are classified into three different zones to indicate different consequences. The predicted results were then compared to the actual data. The result shows that 4 out of 5 companies are predicted correctly with approximately 80% accuracy. The results are corroborated by 292 companies maintaining a 75% accuracy. Conclusively, the computed outcome from the case study suggests that Grover's model effectively predicts bankruptcy risk with an accuracy ranging between 75% and 80%.

Bankruptcy Prediction, Financial Distress and Corporate Life Cycle: Case Study of Central European Enterprises

Bankruptcy Prediction, Financial Distress and Corporate Life Cycle: Case Study of Central European Enterprises

Investigating the Factors Predicting the Financial Distress of Firms: A Study Based on Pakistani Firms

This study tries to investigate the determinants of financial distress or bankruptcy prediction of the non-financial firms listed on the Pakistan Stock Exchange 100 Index. The study has taken a sample of 21 non-financial firms representing the cement, sugar, and textile sectors for analysis in the period spanning from 2014 to 2023. Altman's Z-score is considered as proxy for financial distress. This study applies the Panel Least Squares (PLS) approach to test the hypotheses besides the Fixed Effect (FEM) and Random Effect (REM) models. A comparative assessment of PLS, FEM, and REM was also performed in comparing the model performance. The results indicate that the liquidity, efficiency, and solvency ratios are positively associated with the financial distress measured using the Altman Z-score. On the other hand, leverage and firm size show a very important negative relationship with financial distress; this may further mean that even large firms may become distressed due to other contributing factors. Profitability, as represented by profit margins, does not show any significant impact on financial distress in this study. The study underscores the importance of using bankruptcy prediction models to enable firms to identify likely financial distresses early. Companies also need to look forward toward other factors that may cause distress and focus on if cash flow and debt management practices need improvement to meet their financial obligations efficiently.

Comprehensive Evaluation of Bankruptcy Prediction in Taiwanese Firms Using Multiple Machine Learning Models

Comprehensive Evaluation of Bankruptcy Prediction in Taiwanese Firms Using Multiple Machine Learning Models

An experimental survey of imbalanced learning algorithms for bankruptcy prediction

Information about imminent bankruptcy is crucial for financial institutions, decision-making managers, and state agencies. Since bankruptcy prediction is a prevalent research topic, many new methods have been continuously proposed. Bankruptcy prediction is frequently approached as a binary classification task. Since bankruptcy datasets are inherently imbalanced, bankruptcy classification is usually performed using class imbalance learning methods. The nature of these methods is very diverse, but they can usually be categorized as ensemble, cost-sensitive, sampling, and hybrid methods. In this paper, we provide a comprehensive experimental comparison of 45 methods. These methods were selected because they cover the approaches and algorithms frequently employed for bankruptcy prediction and imbalanced learning. Extensive experiments on 15 publicly available datasets with different imbalance ratios showed that the methods based on a combination of ensemble learning and undersampling are able to handle data imbalance and achieve the best results for bankruptcy classification.

European professional football in the light of bankruptcy prediction models: a case for the insolvency administrator?

ABSTRACT Purpose/Rationale European elite football often fails to translate its economic success story and revenue growth into financially viable businesses. This paper aims to extend the current research on financial performance in European football, adding a significant but missing pan-European comparison to the research field. It analyzes the bankruptcy likelihood of clubs in four leading European football leagues from 2017–2022. Approach We use Altman’s Z-Score to assess the financial risks of 88 clubs in England (28), Spain (23), Germany (17), and Italy (20) that played in the national top division for at least one season during the period under review. Panel regression is used to assess the impact of five performance variables on Z-Scores. Data were sourced from dissecting clubs’ annual accounts and extracted from other football financial databases. Findings The results show significant financial distress across European football clubs. The analysis emphasizes that the COVID pandemic has exacerbated clubs’ financial challenges, but not caused them. However, the absence of spectators due to COVID restrictions did aggravate the situation, as stadium utilization positively correlated to Z-Scores. Fixed-effects panel regressions show that a balanced wages-to-revenue ratio is a critical performance metric for improving Z-Scores. Practical Implications European football needs collective governance efforts to enforce stricter financial discipline post-COVID and remedy its financial situation. Research Contribution The findings have significant implications for policymakers, club managers, and investors, highlighting the financial risks across Europe’s the four leading football leagues. Originality This first pan-European analysis compares clubs across different geographies, thus extending existing country-view work/approaches.

Advancing bankruptcy prediction: a study on an improved rime optimization algorithm and its application in feature selection

Advancing bankruptcy prediction: a study on an improved rime optimization algorithm and its application in feature selection

Challenges of Artificial Intelligence for the Prevention and Identification of Bankruptcy Risk in Financial Institutions: A Systematic Review

The identification and prediction of financial bankruptcy has gained relevance due to its impact on economic and financial stability. This study performs a systematic review of artificial intelligence (AI) models used in bankruptcy prediction, evaluating their performance and relevance using the PRISMA and PICOC frameworks. Traditional models such as random forest, logistic regression, KNN, and neural networks are analyzed, along with advanced techniques such as Extreme Gradient Boosting (XGBoost), convolutional neural networks (CNN), long short-term memory (LSTM), hybrid models, and ensemble methods such as bagging and boosting. The findings highlight that, although traditional models are useful for their simplicity and low computational cost, advanced techniques such as LSTM and XGBoost stand out for their high accuracy, sometimes exceeding 99%. However, these techniques present significant challenges, such as the need for large volumes of data and high computational resources. This paper identifies strengths and limitations of these approaches and analyses their practical implications, highlighting the superiority of AI in terms of accuracy, timeliness, and early detection compared to traditional financial ratios, which remain essential tools. In conclusion, the review proposes approaches that integrate scalability and practicality, offering predictive solutions tailored to real financial contexts with limited resources.

Bankruptcy Prediction for US Businesses: Leveraging Machine Learning for Financial Stability

The economic ramifications due to the bankruptcy of businesses in the USA are exponentially huge and multi-dimensional. Starting from small businesses to huge and large-scale businesses, all declare bankruptcy every year, leading to massive sacking, reduced consumer confidence, and consequently a trickled effect throughout other sectors of the economy. The prime objective of the present study was to devise and execute machine learning techniques to predict bankruptcy in US businesses effectively. This research project intends to develop an efficient understanding of the factors leading to business failures using algorithms that learn from data. For the present study focusing on bankruptcy prediction, we used several datasets to enhance the quality and reliability of forecasts. The major data sources were financial statements, which include balance sheets, income statements, and cash flow statements, providing quantitative measures that enable analysts to perceive the financial health of a firm through various ratios and indicators. Machine learning model selection for the prediction of bankruptcy is based on the evaluation of various algorithms: Logistic Regression, Random Forest, Gradient, and Boosting. The models were evaluated against a set of overall metrics: accuracy, precision, recall, F1-score, and ROC-AUC. Random Forest and XG-Boost resulted in marginally better scores across all metrics as compared to Logistic Regression. Predictive insights determined from bankruptcy risk models give rise to valuable interpretations for decision-makers. An organization in the USA can, from model prediction analysis, identify firms that show a high risk of going into bankruptcy and thus enable appropriate interventions in time. Machine learning-driven bankruptcy prediction undoubtedly assists in integrating better risk management policies and procedures in financial institutions. Similarly, by using complex algorithms for pattern identification in historical data, an institution will go deeper in identifying patterns constituting distress in companies

A machine learning and ANOVA-based approach to model financial predictors for corporate failure in imbalanced dataset: The case of Taiwan

This study meticulously explores the crucial elements precipitating corporate failures in Taiwan during the decade from 1999 to 2009. It proposes a new methodology, combining ANOVA and tuning the parameters of the classification so that its functional form describes the data best. Our analysis reveals the ten paramount factors, including Return on Capital ROA(C) before interest and depreciation, debt ratio percentage, consistent EPS across the last four seasons, Retained Earnings to Total Assets, Working Capital to Total Assets, dependency on borrowing, ratio of Current Liability to Assets, Net Value Per Share (B), the ratio of Working Capital to Equity, and the Liability-Assets Flag. This dual approach enables a more precise identification of the most instrumental variables in leading Taiwanese firms to bankruptcy based only on financial rather than including corporate governance variable. By employing a classification methodology adept at addressing class imbalance, we substantiate the significant influence these factors had on the incidence of bankruptcy among Taiwanese companies that rely solely on financial parameters. Thus, our methodology streamlines variable selection from 95 to 10 critical factors, improving bankruptcy prediction accuracy and outperforming Liang’s 2016 results.

Analisis Prediksi Kebangkrutan Menggunakan Model Altman Z-Score dan Zmijewski pada Perusahaan Multifinance

Bankruptcy is a problem that companies must be aware of. If the company experiences bankruptcy, then the company is experiencing business failure. One of the ways that companies do to detect and prevent this from happening is by monitoring financial performance using financial statement analysis. To be able to predict company bankruptcy usually uses several different models, in this study using the Altman and Zmijewski models. The purpose of this study was to analyze the prediction of bankruptcy in multifinance sector companies using the Altman and Zmijewski models. The data used in this study are secondary data in the form of company financial reports published on the Indonesia Stock Exchange website. The population in this study were all multifinance companies listed on the IDX, totaling 14 companies. The samples used in this study amounted to 14 companies using the census method or total sampling. The results of this study indicate that there are multifinance sector companies that are predicted to go bankrupt using the Altman Z-Score Model, there are multifinance sector companies that are predicted to go bankrupt using the Zmijewski Model, and there are significant differences between the Altman Z-Score Model and the Zmijewski Model in predicting bankruptcy in multifinance sector companies.

Comparison Of Corporate Bankruptcy Predictions Based On The Altman Model And Springate Approach

This study was conducted to determine the financial performance of technology sector companies from the analysis of bankruptcy prediction using the Altman Z-Score and Springate S-Score methods and compare the accuracy and type error of the two methods in the 2021-2023 period. This study uses secondary data obtained from the Indonesia Stock Exchange using quantitative descriptive analysis techniques. The determination of the sample in this study uses a nonprobability sampling method, precisely with a purposive sampling technique. The results of this study show that based on the Altman Z-Score method, there are 4 technology sector companies that have the potential to experience bankruptcy for 3 consecutive years, including ATIC, DIGI, CASH, ENVY. Based on the Springate S-Score method, it is indicated that 4 companies have the potential to go bankrupt for 3 consecutive years, including DIGI, CASH, ENVY, CYBR. The accuracy and type error rate obtained for three years in technology sector companies based on the Springate S-Score method resulted in a higher accuracy of 72% in 2022 with lower type errors compared to the Altman Z-Score method. Keywords: Technology Sector, Bankruptcy, Altman Z-Score, Springate S-Score

Significant Features with M-Polar Neutrosophic Topological Spaces and Grey Wolf Optimization Algorithm for Bankruptcy Prediction Model

Significant Features with M-Polar Neutrosophic Topological Spaces and Grey Wolf Optimization Algorithm for Bankruptcy Prediction Model

A Comparative Analysis of Bankruptcy Prediction Models for the Korean Shipping Companies and Extraction of Key Risk Factors

A Comparative Analysis of Bankruptcy Prediction Models for the Korean Shipping Companies and Extraction of Key Risk Factors

Optimization of XGBoost Bankruptcy Prediction Based on Four-Vector Optimization Algorithm

In this paper, the XGBoost model is optimized based on the four-vector optimization algorithm to realize the effective prediction of enterprise bankruptcy. In the experiment, we adopted this advanced algorithm to greatly improve the predictive performance of the model, with the goal of providing a reliable bankruptcy risk assessment tool for enterprises and investors. By training the model, we observed a gradual decline in the fitness curve from 0.029 to 0.026. This trend indicates that with the increase of the number of training iterations, the performance of the model is constantly improving. The gradual slowing of fitness decline suggests that the model may have approached the optimal solution or become stable, which provides confidence for subsequent applications. In terms of specific prediction effect, the confusion matrix of the training set shows that the accuracy of the model is as high as 100%, while the accuracy of the test set is as high as 97.33%. Such high accuracy not only reflects the model's excellent ability on the seen data, but also shows that the model has good generalization ability and can maintain efficient performance on the unseen data. In addition, the performance indicator chart shows the trend of different performance indicators, including FM, J, Q, CA, DAUC, SE, SP. From the figure, we can see that the values of FM and J remain around 1.5, indicating that the model performs well in some ways. However, DAUC and SE showed less than ideal results, close to zero or negative, which means that on some performance indicators, the model still needs further optimization and improvement. To sum up, the research in this paper not only provides an effective tool for corporate bankruptcy prediction, but also provides data support for investors and managers in the decision-making process. By accurately predicting the risk of bankruptcy, relevant parties can take preventive measures to effectively reduce investment risks and improve decision-making efficiency. Therefore, improving the accuracy of corporate bankruptcy prediction can not only promote the sustainable development of enterprises, but also provide protection for the healthy operation of the economy in a larger scope.

Construction of Discrimination Models in Prediction of Bankruptcy if Polish Non-Public Enterprises

The purpose of the article. The aim of this study is to predict bankruptcy among Polish non-financial firms by constructing discriminant models and comparing the outcomes with prognostic models developed by other Polish scholars. Utilizing financial data from 2017–2021 for 416 companies across the trade, production, and service sectors, this research strives to devise the most effective model for classifying entities into two groups. Methodology. The study employed a discriminant function, a statistical method enabling the classification of objects based on several explanatory variables simultaneously. Two methods for selecting independent variables for the discriminant function were compared using group mean equality tests and Hellwig's method. Additionally, two techniques of winsorization were applied to minimize the impact of outliers on the study results. Results of the research. The study’s findings underscore the importance of operational profitability relative to total assets and the logarithm of total assets as key variables in bankruptcy prediction models. Results confirm the significance of industry specificity on the models' classification accuracy. The use of different methods for selecting independent variables for models and winsorization directly impacts classification efficacy. A comparative analysis with models from selected Polish researchers reveals that the models developed in this study achieved a higher level of effectiveness than existing models in terms of classification accuracy.

Comparation Of Bankruptcy Prediction At Retail Companies In Indonesia Using Altman, Zmijewski and Springate Methods

The Indonesian retail sector, a significant contributor to the nation's GDP, faces challenges due to digital transformation, shifting consumer behavior, and increased competition from e-commerce platforms, leading to potential bankruptcy risks among traditional retailers. This study aims to compare the Altman, Zmijewski, and Springate models in predicting bankruptcy of retail companies listed on the Indonesia Stock Exchange from 2019 to 2023. Using financial data from 11 retail companies, the study calculated bankruptcy predictions using the three models and performed statistical tests (Kolmogorov-Smirnov and Kruskal-Wallis tests) to compare their predictive accuracy. The results indicated that the Zmijewski model had the highest prediction accuracy at 90.91%, followed by the Springate model at 81.82%, and the Altman model at 45.45%. The Zmijewski model predicted one company as potentially bankrupt, while the other models provided varying predictions. The study concludes that the Zmijewski model is highest prediction rate in predicting bankruptcy in Indonesian retail companies. This has implications for investors and stakeholders in assessing financial health and making informed decisions

The dilemma of accuracy in bankruptcy prediction: a new approach using explainable AI techniques to predict corporate crises

Purpose Our aim is to develop a highly precise corporate crisis prediction model that surpasses previous versions, rooted in the forefront of technological advancements.Design/methodology/approach Artificial Intelligence (AI) for corporate default prediction with a novel approach based on a mix of techniques, enabling it to achieve a higher accuracy. We investigated models with sequence lengths that were both fixed and variable, and we chose the best variable sequence length model.Findings Our findings demonstrate that the artificial techniques implemented lead to very high accuracy in predicting business crises compared to previous research efforts, even those utilising long-time sequences or a high volume of observations.Research limitations/implications We highlight the key variables with a higher predictive power that need monitoring to prevent business crises. We also aim to open a new avenue of research that, starting from the use of these techniques and our results, can implement models incorporating non-accounting variables to prevent business crises.Practical implications We provide a model/tool that assesses a possible business crisis in advance through a monitoring and alert system. Policymakers can use our research’s output as a tool to combine with current credit-scoring systems and to assess the effectiveness of the new corporate crisis reforms that are upcoming in many European countries. The results of our research can be useful also to banks, public entities, and consulting firms that interact with companies and are interested in the evaluation of a firm’s financial health and stability.Originality/value Our innovative work leverages cutting-edge methodologies such as deep Recurrent Neural Networks and explainable AI. This choice is driven by the rapid evolution of AI techniques in practical application.

Bankruptcy Prediction Analysis of General Insurance Companies Using the Ohlson Model

General insurance companies play an important role in maintaining economic stability by transferring financial risks from individuals and companies to insurance companies. However, insurance companies are not immune to the risk of bankruptcy that can arise due to factors such as inability to manage claims, premium fluctuations, and insufficient capital. Early detection of potential bankruptcy becomes very important to prevent greater losses. This study aims to analyze the prediction of bankruptcy in general insurance companies in Indonesia using the Ohlson Model. The Ohlson model is based on logistic regression, taking into account several financial variables such as leverage, profitability, and company size to estimate the probability of bankruptcy. The results of the study are expected to provide insights for insurance company management and regulators in identifying bankruptcy risks and taking appropriate preventive measures. In addition, this study contributes to enriching the literature related to the application of bankruptcy prediction models in the context of the insurance industry in emerging markets. From the analysis, it was found that out of 13 general insurance companies listed on the Indonesia Stock Exchange (IDX), the Ohlson value for all companies is below 0.38, which indicates that the sampled companies still have fairly good financial stability. The research results are expected to provide insights for insurance company management and regulators in identifying bankruptcy risks and taking appropriate preventive measures. In addition, this study contributes to enriching the literature related to the application of bankruptcy prediction models in the context of the insurance industry in emerging markets.