Combining Sampling Methods, Cost‐Sensitive Learning, and Ensemble Techniques for Highly Class‐Imbalanced Financial Distress Prediction

Previous studies used constant cut‐off indicator to distinguish distressed firms from non‐distressed ones in the one‐stage prediction models. However, distressed cut‐off indicator must shift according to economic prosperity, rather than remains fixed all the time. This study focuses on Taiwanese listed firms and develops financial distress prediction models based upon the two‐stage method. First, this study employs the firm‐specific financial ratio and market factors to measure the probability of financial distress based on the discrete‐time hazard models. Second, this paper further focuses on macroeconomic factors and applies rating transition matrix approach to determine the distressed cut‐off indicator. The prediction models are developed by using the training sample from 1987 to 2004, and their levels of accuracy are compared with the test sample from 2005 to 2007. As for the one‐stage prediction model, the model in incorporation with macroeconomic factors does not perform better than that without macroeconomic factors. This suggests that the accuracy is not improved for one‐stage models which pool the firm‐specific and macroeconomic factors together. In regards to the two stage models, the negative credit cycle index implies the worse economic status during the test period, so the distressed cut‐off point is adjusted to increase based on such negative credit cycle index. After the two‐stage models employ such adjusted cut‐off point to discriminate the distressed firms from non‐distressed ones, their error of misclassification becomes lower than that of one‐stage ones. The two‐stage models presented in this paper have incremental usefulness in predicting financial distress.

PurposeThis study examines the effect of non-financial variables related to governance on the accuracy of financial distress prediction among Egyptian listed small and medium-sized enterprises (SMEs), by using the logistic regression technique.Design/methodology/approachThis study used a sample of 24 Egyptian-listed SMEs in each year, totaling 120 firm observations, of which 25 were classified distressed and 95 of them non-distressed between 2014 and 2018. The variables for the study included five financial variables and thirteen non-financial variables related to governance. The models were developed using financial variables alone as well as combining financial and non-financial variables related to governance.FindingsThe results showed that the model with financial variables had a prediction accuracy of 91.7% , whereas models with a combination of financial and non-financial variables related to governance predict with comparatively better accuracy of 92.7 and 93.6% .Research limitations/implicationsAlthough the results seem to be conclusive, it could be noted that the non-distressed sample was not paired with the distressed sample. Other studies showed that paired samples increase the financial distress prediction rate. Furthermore, due to the small sample size, this study was unable to create a hold-out sub-sample for the accuracy test.Practical implicationsThe proposed distress prediction model for SMEs is effective for stakeholders, including banks and other financial institutions, in the assessment of the credit risk of SMEs. Using such a model, they could better identify SMEs with a higher risk of failure in their lending decisions. Moreover, SME managers' could be interested in using such models as a tool for planning corrective action, in addition to planning and controlling current operations to avoid financial failure in the future.Originality/valueThis study contributes to financial distress prediction literature in different ways. First, few studies were conducted in the area of financial distress among SMEs. Second, neither of these studies was conducted within the Egyptian context, nor any of them had used non-financial variables related to governance in the prediction of financial distress among SMEs.

In recent years, financial distress prediction (FDP), also known as corporate failure prediction or bankruptcy prediction, has gained significant importance due to its impact on organizations, especially during unexpected events like pandemics and wars. Machine learning (ML) models have emerged as innovative and essential tools in predicting financial distress, leveraging the ever‐increasing volume of databases and computing power. This study utilizes bibliographic techniques to contribute to the field's literature review to address the disorganized nature of the existing literature on FDP, reduce confusion, and provide clarity to domain researchers. These techniques enable identifying the progress of articles published over the years, influential authors, and highly cited articles. Additionally, the study examines crucial aspects of data preprocessing, such as missing data, imbalanced data, feature selection, and outliers, as they significantly impact the robustness and performance of ML models. Furthermore, it discusses essential models employed in FDP, focusing on recent advancements that represent promising trends. In conclusion, this study contributes to the field by uncovering novel trends and proposing possible directions for advancing FDP research. These findings will guide researchers, practitioners, and stakeholders in their quest for improved prediction and decision‐making in financial distress.

Accurately predicting financial distress is a critical issue in financial decision-making. Financial distress must be detected as early as possible as an important determining factor in decision-making for internal companies and financial institutions related to financing or loan decisions. Various studies on financial distress prediction in Indonesia have been carried out, ranging from traditional statistical approaches to machine learning. However, the performance of the two methods is still not optimal. Therefore, this study tries to develop machine learning techniques by combining cluster analysis and classification in a hybrid model to improve the prediction model’s performance. The case study adopted in this study is the prediction of financial distress in non-financial companies listed on the IDX from 2018-2021 by combining k-means clustering and Support Vector Machine. The analysis results show that the hybrid classifier has an accuracy value of 92.7%, which is higher than the accuracy of the single classifier, which is 88.6%.

Survey, classification and critical analysis of the literature on corporate bankruptcy and financial distress prediction

Financial distress prediction (FDP) has been a subject of extensive and ongoing research because of its significance in both internal and external components of enterprises including investors and creditors. Financial institutions must to be able to foresee financial difficulty in order to allow them for evaluating the financial health of businesses and individuals. Data pre-processing techniques have been found to increase the efficacy of prediction models, and many research consider feature selection as a pre-processing step before creating the models. The creation of efficient feature selection algorithms is one of the main challenges facing FDP. In this study, we present a hybrid methodology for predicting financial distress using a Multi-Layer Perceptron and Genetic Algorithm (MLP_GA) model with boruta automated feature selection. The proposed model is designed on genetic algorithm- based tuning of the crucial MLP hyperparameters, including Network depth, Dense layer activation function, Network width, and Network optimizer for a reliable prediction. This paper investigates how boruta algorithm based feature selection method improve the accuracy of our MLP_GA algorithm. We access the FDP performance utilizing samples of enterprises based in MENA area. Resampling with k-fold evaluation metrics is employed in the experiments. The experimental results indicate that the adoption of the boruta automated feature selection method has significantly enhanced the prediction performance and accuracy of the FDP model.

Financial distress is a common global phenomenon among the corporate entities. Locally, there is overwhelming evidence of firms that have undertaken financial restructuring, delisted from the exchange market, gone into receivership and subsequently liquidated on account of financial distress. This study therefore set out to examine the way in which management of working capital affects financial distress of non-financial firms listed at the Nairobi Securities exchange. In fulfilling this objective, the study sought to establish how cash management, inventory management and accounts receivables management effects financial distress of non-financial firms listed at Nairobi Securities Exchange. The free cash flows theory, Precautionary motive theory, financing advantage theory and liquidity theory formed the theoretical basis of the study. The study adopted longitudinal research design and collected secondary data over ten years period (2009-2018) from a census of the 40 non-financial firms listed in Nairobi Securities exchange. Descriptive statistical analysis was used to obtain the initial overview of the data collected. Inferential statistical analysis was undertaken using the F and t-tests at 95% confidence level. The study found that cash management had a positive and significant effect on the firms’ distress index. Further, the study revealed that inventory holding period was negatively and significantly related to the firms’ financial distress index. The study also showed that suppliers’ payment period had a positive and significant effect on financial distress indicator. The study however depicted a negative but insignificant relationship between receivables period and financial distress. The study recommended that the management of non-financial listed firms should ensure appropriate management of working capital components in order to guard against instances of corporate financial distress 
 
 JEL: O15; J24; L20
 
 <p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/edu_01/0778/a.php" alt="Hit counter" /></p>

Predicting financial difficulties provides valuable early warnings for companies, making it a significant area of study. Additionally, creditors and investors stand to benefit from being able to foresee financial distress. Machine learning (ML) models have emerged as essential tools for predicting financial trouble, leveraging expanding databases and processing capacity. This study undertakes a methodical examination of the literature to address the existing fragmentation in research on machine learning trends in financial distress prediction, while exploring their implications for decision-making effectiveness and sustainability outcomes. The originality of this paper is that no previous study has described all the methods in depth and their role in promoting sustainability, which makes a novel contribution to this study. It also employs content analysis to identify crucial themes and examines significant aspects of data preprocessing techniques. Furthermore, the study delves into key models in Financial Distress Prediction (FDP) and recent trends, exploring the advantages and disadvantages of various machine learning techniques to provide guidance for researchers, practitioners, and stakeholders in predicting financial distress and making well-informed decisions. In addition, the study highlights an emerging research gap by emphasizing the challenges and opportunities associated with applying machine learning techniques for financial distress prediction in developing countries, thereby extending the scope for future investigations.

Financial distress prediction remains a central topic in corporate finance and risk management, with extensive research devoted to improving classification accuracy through increasingly sophisticated statistical and machine learning techniques. Nevertheless, the influence of data preparation on predictive performance has received comparatively less systematic attention. This study examines how an economically grounded data-preparation process affects the predictive performance of selected statistical and machine-learning models dedicated to predicting corporate financial distress. Using the chosen financial ratios, generally accepted indicators of corporate financial stability and economic performance, financial distress models are estimated on both raw, unprocessed input data and pre-processed data involving the exclusion of economically implausible accounting values, treatment of missing observations, and class balancing. In light of the above, the study adopts a structured methodological approach to assess the predictive performance of selected classification models, namely decision tree algorithms (CART, CHAID, and C5.0), artificial neural networks (ANNs), logistic regression (LR), and linear discriminant analysis (DA), using confusion-matrix–based evaluation and a comprehensive set of evaluation measures. The results suggest that the process of input data preparation is a critical factor, significantly improving the predictive performance of financial distress prediction models across most modelling techniques employed. The most pronounced gains are observed in decision tree models. ANNs also demonstrate marked improvement after input data preparation, whereas LR benefits more moderately, and linear DA remains limited despite preprocessing. The average gain in accuracy across all six modelling techniques, calculated as the difference between pre-processed and raw performance for each method and averaged across methods, was approximately 15.6 percentage points, with specificity improving by approximately 26.9 percentage points on average, amounting to roughly half the performance variation attributable to algorithm choice, which underscores that data preparation is a primary determinant of model reliability alongside algorithm selection. A step-level detailed analysis further shows that missing value imputation is the dominant driver of improvement for tree-based models, while class balancing contributes most for ANNs and logistic regression. The findings highlight that reliable financial distress prediction depends not only on technique selection but also on the consistency and economic plausibility of the input data, underscoring the central role of structured data preparation in developing robust early-warning models.

The prediction for listed companies’ financial distress by using multiple prediction methods with rough set and Dempster–Shafer evidence theory

Aim/Purpose: This paper aims to empirically quantify the financial distress caused by the COVID-19 pandemic on companies listed on Amman Stock Exchange (ASE). The paper also aims to identify the most important predictors of financial distress pre- and mid-pandemic. Background: The COVID-19 pandemic has had a huge toll, not only on human lives but also on many businesses. This provided the impetus to assess the impact of the pandemic on the financial status of Jordanian companies. Methodology: The initial sample comprised 165 companies, which was cleansed and reduced to 84 companies as per data availability. Financial data pertaining to the 84 companies were collected over a two-year period, 2019 and 2020, to empirically quantify the impact of the pandemic on companies in the dataset. Two approaches were employed. The first approach involved using Multiple Discriminant Analysis (MDA) based on Altman’s (1968) model to obtain the Z-score of each company over the investigation period. The second approach involved developing models using Artificial Neural Networks (ANNs) with 15 standard financial ratios to find out the most important variables in predicting financial distress and create an accurate Financial Distress Prediction (FDP) model. Contribution: This research contributes by providing a better understanding of how financial distress predictors perform during dynamic and risky times. The research confirmed that in spite of the negative impact of COVID-19 on the financial health of companies, the main predictors of financial distress remained relatively steadfast. This indicates that standard financial distress predictors can be regarded as being impervious to extraneous financial and/or health calamities. Findings: Results using MDA indicated that more than 63% of companies in the dataset have a lower Z-score in 2020 when compared to 2019. There was also an 8% increase in distressed companies in 2020, and around 6% of companies came to be no longer healthy. As for the models built using ANNs, results show that the most important variable in predicting financial distress is the Return on Capital. The predictive accuracy for the 2019 and 2020 models measured using the area under the Receiver Operating Characteristic (ROC) graph was 87.5% and 97.6%, respectively. Recommendations for Practitioners: Decision makers and top management are encouraged to focus on the identified highly liquid ratios to make thoughtful decisions and initiate preemptive actions to avoid organizational failure. Recommendation for Researchers: This research can be considered a stepping stone to investigating the impact of COVID-19 on the financial status of companies. Researchers are recommended to replicate the methods used in this research across various business sectors to understand the financial dynamics of companies during uncertain times. Impact on Society: Stakeholders in Jordanian-listed companies should concentrate on the list of most important predictors of financial distress as presented in this study. Future Research: Future research may focus on expanding the scope of this study by including other geographical locations to check for the generalisability of the results. Future research may also include post-COVID-19 data to check for changes in results.

In real-world anomaly detection tasks such as Credit Card Fraud Detection, Cancer Patients Detection, Phishing Website Detection, etc., the training datasets often suffer from skewed class distribution. But the traditional Machine Learning (ML) classification algorithms assume balanced class distribution and equal misclassification costs. As a result, when class-imbalanced data are presented to the traditional ML algorithms they tend to produce biased and inaccurate predictive ML models. In this study, we propose four novel Phishing Website Classification models namely, SMOTEENN-XGB, SMOTEENN-RF, SMOTEENN-LR, and SMOTEENN-SVM by combining SMOTEENN (SMOTE + ENN) hybrid sampling technique with eXtreme Gradient Boosting (XGB), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) classifiers respectively. We propose the use of SMOTEENN hybrid sampling as the novel approach to address the problem of class imbalance in Phishing Website datasets prior to building classification models. To the best of our knowledge and belief, our novel proposed four models SMOTEENN-XGB, SMOTEENN-RF, SMOTEEEN-LR, and SMOTEENN-SVM for Phishing Website Detection based on SMOTEENN hybrid sampling approach have not been published in the existing studies as of now.

PurposeThis study aims to analyze the effect of cognitive load and social value orientation on managers’ preferences when they face with two types of restructuring choices in financially distressed firms: the first belonging to the family of organizational restructuring (massive layoffs) and the second to the family of financial restructuring (debt increases).Design/methodology/approachThe authors investigate experimentally the impact of managers’ cognitive load and social value orientation on the decision to restructure leveraged buyout (LBO) firms in financial distress by using either massive layoffs or debt increases.FindingsBy investigating the impact of managers’ cognitive load and social value orientation on the restructuring decision of an LBO firm in financial distress, the research reveals that, on average, cognitively loaded managers prefer massive layoffs over increased debt levels. The massive layoffs seemingly provide a relatively easier way to avoid conflict with influential, residual claimants. In contrast, social value–oriented managers actively avoid massive layoffs and prefer to increase debt.Research limitations/implicationsThese results imply that the performance mechanisms emphasized to improve agency relations, for example, in LBOs, have their own limitations during periods of financial distress. This study shows that one of these limits is related to cognitive distortions and personality traits.Originality/valueIn this research, the originality lies in understanding how managers’ internal factors affect their restructuring decision-making, in the case of LBO firms in financial distress.

This study claims that two key points must be addressed in order to increase the success rate and robustness of financial distress prediction: firstly, a simplified financial prediction indicator system must be established based on a correct understanding and clearly distinguish about the concept of financial distress. Secondly, we establish a reasonable and plausible financial distress prediction model. It has been established that machine learning models can enhance financial distress prediction models' predictive power to some degree, bu the outcomes of these models vary. This is mostly due to a lack of knowledge about financial distress and a poor choice of indicators for predicting financial distress. This paper makes the case that there are three distinct, dynamic phases of financial distress. The financial strain stage is the initial phase. The financial distress at this stage is reflected in the financial indicators, as there are now some challenges with loan repayment due to the slowdown in the main business's revenue growth rate, the slowdown in operating cash flow, and the drop in the current ratio. The second stage can be called the financial crisis stage. The enterprise's ability to generate income is still declining, along with the quality of revenue and turnover rate. It’s worth noting that gearing ratio, which will be approaching 50% or above at this stage. The third stage of financial distress is distress situation. In this stage, the firm’s balance sheet structure keeps getting worse, and the gearing ratio keeps rising at an unprecedented rate. In the meantime, with growing revenue-cost-expense ratios and negative profits, profitability is still declining. At this stage, at least two of the three cash flows from financing, investing, and operating operations were negative in terms of cash flow.

This study thoroughly examined the use of different machine learning models to predict financial distress in Indonesian companies by utilizing the Financial Ratio dataset collected from the Indonesia Stock Exchange (IDX), which includes financial indicators from various companies across multiple industries spanning a decade. By partitioning the data into training and test sets and utilizing SMOTE and RUS approaches, the issue of class imbalances was effectively managed, guaranteeing the dependability and impartiality of the model’s training and assessment. Creating first models was crucial in establishing a benchmark for performance measurements. Various models, including Decision Trees, XGBoost, Random Forest, LSTM, and Support Vector Machine (SVM) were assessed. The ensemble models, including XGBoost and Random Forest, showed better performance when combined with SMOTE. The findings of this research validate the efficacy of ensemble methods in forecasting financial distress. Specifically, the XGBClassifier and Random Forest Classifier demonstrate dependable and resilient performance. The feature importance analysis revealed the significance of financial indicators. Interest_coverage and operating_margin, for instance, were crucial for the predictive capabilities of the models. Both companies and regulators can utilize the findings of this investigation. To forecast financial distress, the XGB classifier and the Random Forest classifier could be employed. In addition, it is important for them to take into account the interest coverage ratio and operating margin ratio, as these finansial ratios play a critical role in assessing their performance. The findings of this research confirm the effectiveness of ensemble methods in financial distress prediction. The XGBClassifier and RandomForestClassifier demonstrate reliable and robust performance. Feature importance analysis highlights the significance of financial indicators, such as interest coverage ratio and operating margin ratio, which are crucial to the predictive ability of the models. These findings can be utilized by companies and regulators to predict financial distress.