Corporate Bankruptcy Prediction Research Articles

Predicting Corporate Bankruptcy: A Comparative Analysis of Machine Learning Models

Predicting Corporate Bankruptcy: A Comparative Analysis of Machine Learning Models

Corporate Bankruptcy Prediction Models: A Comparative Study for the Construction Sector in Greece

This study focuses on testing the efficiency of alternative bankruptcy prediction models (Altman, Ohlson, Zmijewski) and on assessing the possible reasons that led to the confirmation or not of the prevailing model. Data from financial statements of listed (Greek) construction companies before the economic crisis were utilized. The results showed that Altman’s main predictive model as well as the revised models have low overall predictability for all three years before bankruptcy.

A Bibliometric Analysis of Corporate Bankruptcy Prediction Research: Trends and Insights

A Bibliometric Analysis of Corporate Bankruptcy Prediction Research: Trends and Insights

Are Bankruptcy Prediction Models Effective for Predicting Ratings? An Empirical Investigation of Altman’s Model in Indian Corporate Sector

The alternative method for predicting credit ratings by employing bankruptcy prediction models is the main subject of the present article. Rating agencies utilize an extremely intricate rating methodology, which differs from agency to agency, and the ratings given by agencies are less reliable in anticipating short-term defaults and react slowly to credit events, as demonstrated by the failure of agencies during the crisis of 2008. Both credit rating assignment and bankruptcy prediction focused on the probability of default, but the methodology used by the two varies significantly: corporate bankruptcy prediction models predict the future default probability of a company as a whole, whereas credit rating only access the default risk associated with a specific instrument. Therefore, it is preferable to employ bankruptcy prediction algorithms to access a company’s future ratings. The present study used Altman’s framework in order to assess whether bankruptcy prediction models can be applied to access a company’s future ratings in advance or not and found that it can be used as an alternative method to predict the future credit ratings, which can be used by investors and banks to make better investment and lending decisions.

FAKTOR-FAKTOR YANG MEMPENGARUHI PREDIKSI KEBANGKRUTAN PERUSAHAAN PADA PERUSAHAAN MANUFAKTUR

This research aims to examine the effect of cash flow, profitability, and leverage on the predicition of corporate bankruptcy in manufacturing companies listed on the Indonesia Stock Exchange for the period 2018-2020. This study used a purposive samnpling method and obtained 19 companies and total of samples in this study were 57 samples. The method in this study used logistic regression analytics with Eviews12 and Microsoft Excel. The results of this study shows that cash flow, profitability, and leverage has an insignificant effect on prediction of corporate bankruptcy. Then, the cash flow variables. profitability, and leverage simultaneously has an effect on prediction of corporate bankruptcy.

Bankruptcy prediction with low-quality financial information

The corporate bankruptcy prediction literature has traditionally relied on data from public, audited companies. However, the vast majority of firms worldwide are privately-held and lack the same level of scrutiny over their financial statements. As a result, these businesses usually produce less accurate and transparent accounting reports. Our research problem is to address this gap: how stakeholders deal with these less reliable information? Using a novel dataset of 503 private firms that filed for reorganization in Brazil between 2007 and 2020, we found that financial ratios had a significantly lesser effect on explaining default and bankruptcy than what previous research suggested, due in part to the lower information content in the accounting statements within our database. Instead, lenders seem to focus on harder-to-conceal variables, such as collateralizable assets, as well as on institutional factors, like proxies of financial statement quality. There is also concerning evidence that specialized attorneys can ”work the system” in favor of distressed companies regardless of their financial fundamentals. Additionally, we found that machine learning models outperformed traditional statistical ones in different sorts of metrics, corroborating the literature on the superior performance of non-linear approaches on datasets having synergistic causality among its features.

Prediction of Corporate Bankruptcy using Machine Learning

Over the past few decades, corporate bankruptcy has become a source of concern for various business stakeholders such as investors and management. It has also piqued the interest of researchers worldwide. Because there are so many variables that contribute to bankruptcy, it is insufficient to rely solely on a single predictive model; instead, the difficulty is in pinpointing the crucial elements that carry the greatest weight. The significant class imbalance in the data is another significant obstacle that impairs the model's functionality. While many methods, such as Decision Trees, Support Vector Machines, Neural Networks, etc., have been studied before with various pre-processing approaches, we advance this research by applying a novel combination of a Random Forest feature selection technique and SMOTEENN hybrid resampling technique on Polish Bankruptcy dataset. We next apply four classifiers to the altered data: Random Forest, Decision Tree, KNN, and AdaBoost, and assess the effectiveness of each model. As a result of our suggested approach, the Random Forest classifier had the highest accuracy of 89%, while the AdaBoost model as a whole outperformed it with a Recall of 73% and a Geometric Mean of 80%.

Bankruptcy prediction using machine learning models with the text-based communicative value of annual reports

We investigate whether including the text-based communicative value of annual report increases the predictive power of four machine learning models (Logistic regression, Random Forest, XGBoost, and Support Vector Machine) for corporate bankruptcy prediction using U.S. firm observations from 1994 to 2018. We find that the overall prediction effectiveness of these four models (e.g. accuracy, F1-score, AUCs) significantly improves, especially true in the performance of XGBoost and Random Forest models. In addition, we find that annual report text-based communicative value variables significantly reduce models’ Type II error and keep the Type I error at a relatively small level, especially for the short-term bankruptcy forecast. The results reveal that annual report text-based communicative value effectively mitigates the model misidentification of a non-bankrupt firm as a bankrupt firm. Our results also suggest that annual report text-based communicative value is helpful for bank’s corporate loan underwriting decisions. Finally, our findings still hold when considering different testing periods and random state settings, replacing by another publicly available bankruptcy dataset, and introducing neural network models.

Bankruptcy Prediction Model Development and its Implications on Financial Performance in Slovakia

Abstract Research purpose. Financial distress being a global phenomenon makes it impact firms in all sectors of the economy and predicting corporate bankruptcy has become a crucial issue in economics. At the beginning of the last century, the first studies aimed to predict corporate bankruptcy were published. In Slovakia, however, several prediction models were developed with a significant delay. The main aim of this paper is to develop a model for predicting bankruptcy based on the financial information of 3,783 Slovak enterprises operating in the manufacturing and construction sectors in 2020 and 2021. Design / Methodology / Approach. A prediction model that uses the appropriate financial indicators as predictors may be developed using multiple discriminant analysis. Multiple discriminant analysis is currently used in prediction model development. In this case, financial health is assessed using several variables that are weighted in order to maximise the difference between the average value calculated in the group of prosperous and non-prosperous firms. When developing a bankruptcy prediction model based on multiple discriminant analysis, it is crucial to determine the independent variables used as primary financial health predictors. Findings. Due to the discriminant analysis results, the corporate debt level of the monitored firms may be regarded as appropriate. Despite the fact that the model identified 215 firms in financial distress due to an insufficient debt level, 3,568 out of 3,783 Slovak enterprises operating in the manufacturing and construction sectors did not have any problems with financing their debts. The self-financing ratio was identified in the developed model as the variable with the highest accuracy. Based on the results, the developed model has an overall discriminant ability of 93% since bankruptcy prediction models require strong discriminating abilities to be used in practice. Originality / Value / Practical implications. The principal contribution of the paper is its application of the latest available data, which could help in more accurate financial stability predictions for firms during the current difficult period. Additionally, this is a ground-breaking research study in Slovakia that models the financial health of enterprises in the post-pandemic period.

Predicting Corporate Bankruptcy in JSE-AltX Listed Firms

ABSTRACT: The objective of the study is to predict corporate bankruptcy on the JSE-AltX (Alternative Stock Exchange) listed firms by employing Discriminant Analysis (MDA). Since its inception, the JSE-AltX has listed more than 100 firms and majority of them have migrated to the JSE main board. Motivated by the increasing number of delisted firms both pre and post their migration from JSE-AltX to the JSE main board, this study seeks to investigate the financial distress in the JSE-AltX listed firms. This study use a quantitative method to predict bankruptcy for a sample of 20 JSE-AltX listed firms that belongs to a wide range of industries, over the period from 1 January 2004 until 31 December 2015. Based on the previous literature it is evident that financial ratios play a significant role in the prediction of financial distress. Employing the same set of financial ratios (extracted from annual reports mainly the audited consolidated balance sheets and income statements) that are used in bankruptcy prediction as independent variables, the empirical results show that MDA has a statistically significant power in predicting default risk on the JSE-AltX listed firms. The findings show that the discrimination function is significant at the 5% level of significance. The MDA results reveal that, 7 out of the 20 sample firms are prone to bankruptcy, while the rest are not. Furthermore, the model classifies net profit margin (short-term profitability), current ratio (liquidity) and return on capital invested as the most important financial ratios in distinguishing the successful firms from unsuccessful firms post migration from the JSE-AltX to the JSE main board. Generally, this study results have policy implications which regulatory authorities, investors, employees and lenders will find interesting. Firstly, regulatory authorities can find this research useful as it provides effective review of the firm's financial distress conditions and subsequently signals default risk to various stakeholders. Secondly, this research will not only assist in identify potential default risks, but it will also enable different stakeholders to formulate mechanisms or relevant policies and procedures that will allow them to detect and prevent financial distress.

CORPORATE BANKRUPTCY PREDICTION: A SYSTEMATIC LITERATURE REVIEW AND COMPREHENSIVE BIBLIOMETRIC ANALYSIS

Researchers from all over the world have become more interested in bankruptcy prediction during the past 50 years. The prediction of corporate financial difficulties has been the subject of numerous studies ever since Altman revealed the breakthrough bankruptcy prediction model in 1968. The main aim of this research paper is to describe the fundamental concepts associated with the subject of corporate bankruptcy prediction. By identifying the most relevant research papers, nations, and authors in the Web of Science database, an in-depth review of the publications was performed before the analysis. The bibliometric map was created in the VOS Viewer program using the final search result with all available information. The results of the bibliometric analysis reveal that the keywords bankruptcy prediction and classification are the most closely related keywords using the analysis of citations that frequently occur, and that the USA and China developed the most significant international co-author relationships.

CHOOSING RATIO IN THE FINANCIAL DISTRESS PREDICTION MODEL

Purpose – This study aims to find a financial distress prediction model that is suitable for Indonesian companies.Design/methodology/approach – The sample in this study amounted to 150 data. The research sample was grouped into financial distress and non-financial distress. Research data is sourced from the Indonesia Stock Exchange. Discriminant analysis is used to test data and generate financial distress prediction models for manufacturing companies in Indonesia.Findings – Results show that the financial ratios that contribute to the financial distress prediction model are the ratios of profitability, liquidity, and efficiency.Practical Implications – The resulting model can contribute and as a basis for the development of future studies on relevant, robust, and accurate corporate financial distress early warning systems that will help stakeholders to respond to potential bankruptcies accordingly and on timeOriginality – Research on new models to predict corporate bankruptcy in Indonesia which is a developing country is still rare. Most of the literature still uses the Altman Z-Score model.

Bankruptcy prediction in the post-pandemic period: A case study of Visegrad Group countries

Research background: Effective monitoring of financial health is essential in the financial management of enterprises. Early studies to predict corporate bankruptcy were published at the beginning of the last century. The prediction models were developed with a significant delay even among the Visegrad group countries. Purpose of the article: The primary aim of this study is to create a model for predicting bankruptcy based on the financial information of 20,693 enterprises of all sectors that operated in the Visegrad group countries during the post-pandemic period (2020?2021) and identify significant predictors of bankruptcy. To reduce potential losses to shareholders, investors, and business partners brought on by the financial distress of enterprises, it is possible to use multiple discriminant analysis to build individual prediction models for each Visegrad group country and a complex model for the entire Visegrad group. Methods: A bankruptcy prediction model is developed using multiple discriminant analysis. Based on this model, prosperity is assessed using selected corporate financial indicators, which are assigned weights such that the difference between the average value calculated in the group of prosperous and non-prosperous enterprises is as large as possible. Findings & value added: The created models based on 6?14 financial indicators were developed using different predictor combinations and coefficients. For all Visegrad group countries, the best variable with the best discriminating power was the total indebtedness ratio, which was included in each developed model. These findings can be used also in other Central European countries where the economic development is similar to the analyzed countries. However, sufficient discriminant ability is required for the model to be used in practice, especially in the post-pandemic period, when the financial health and stability of enterprises is threatened by macroeconomic development and the performance and prediction ability of current bankruptcy prediction models may have decreased. Based on the results, the developed models have an overall discriminant ability greater than 88%, which may be relevant for academicians to conduct further empirical studies in this field.

Comparing the Performance of Corporate Bankruptcy Prediction Models Based on Imbalanced Financial Data

Forecasts of corporate defaults are used in various fields across the economy. Several recent studies attempt to forecast corporate bankruptcy using various machine learning techniques. We collected financial information on 13 variables of 1020 companies listed on the KOSPI and KOSDAQ to capture the possibility of corporate bankruptcy. We propose a data processing method for small-sample domestic corporate financial data. We investigate the case of random sampling of non-bankrupt companies versus sampling non-bankrupt companies based on approximate entropy and optimized threshold based on AUC to address the imbalance between the number of bankrupt companies and the number of non-bankrupt companies. We compare the performance measures of corporate bankruptcy prediction models for the small sample data structured in two ways and the full dataset. The experimental results of this study contribute to the selection of an appropriate corporate bankruptcy prediction model.

APPLICABILITY OF BANKRUPTCY PROBABILITY ASSESSMENT MODELS TO FINANCIAL SECTOR COMPANIES

Anticipating the likelihood of bankruptcy is essential for every business. Nevertheless, one of the most critical sectors of the economy is the financial sector. The bankruptcy of a company in the financial industry affects both individuals, businesses, and organizations, negatively impacting the economy. Therefore, it is essential to anticipate the likelihood of bankruptcy of companies in the financial sector and make management decisions to avoid the risk of bankruptcy. There are many methodologies for analyzing and predicting corporate bankruptcy that differ in content, number, and accuracy of measurable indicators, so it is crucial to identify which model is most appropriate for assessing the risk of corporate bankruptcy in a particular sector. Given that the financial sector plays a crucial role in economic development and that the consequences of their bankruptcy are harrowing, the aim is to identify the most sensitive model to the risk of bankruptcy in this sector. To achieve this goal, we reveal the concept of bankruptcy, present the internal and external causes of bankruptcy, and systematize the results of bankruptcy risk research of companies in the financial sector. Bankruptcy probability assessment models are also presented and compared, and their applicability to the financial industry is discussed. An analysis of the literature revealed that the most commonly used models for assessing the bankruptcy risk of companies in the financial sector are the Altman Z Index, the Ohlson O Index, and the Zmijewski X Index. After applying these models in the case of three banks (SEB bank, UAB Medicinos bank, and General Financing bank) using 2021. The Altman Z model found that the most sensitive bankruptcy predictor was the probability of bankruptcy.

Benchmarking machine learning models to predict corporate bankruptcy

Benchmarking machine learning models to predict corporate bankruptcy

Bankruptcy Prediction Analysis in Goods Sector Companies Primary Consumers Listed on The Indonesia Stock Exchange

This study aims to determine there are differences in predicting the bankruptcy of primary consumer goods sector companies listed on the Indonesian stock exchange in 2016-2020 by using the Altman Zscore, Zmijewski and Springate bankruptcy prediction models in predicting corporate bankruptcy and to find out which prediction model is better. accurate in predicting corporate bankruptcy. The number of samples used in this study amounted to 18 companies with five years of observation and sampling technique using purposive sampling method. The data analysis technique used is the Kruskal-Wallis test and the level of accuracy test. The results of this study indicate that there is a significant difference from the results of the Altman Zscore, Zmijewski and Springate bankruptcy prediction models using the Kruskal-Wallish H test. with an accuracy rate of 78%.

Evidential Strategies in Financial Statement Analysis: A Corpus Linguistic Text Mining Approach to Bankruptcy Prediction

The qualitative information of companies’ financial statements provides useful information that can increase the accuracy of bankruptcy prediction models. In this research, a dataset of 924,903 financial statements from 355,704 German companies classified into solvent, financially distressed, and bankrupt companies using the Amadeus database from Bureau van Dijk was examined. The results provide empirical evidence that a corpus linguistic approach implementing evidential strategy analysis towards financial statements helps to distinguish between companies’ financial situations. They show that companies use different approaches and confidence assessments when evaluating their financial statements based on solvency and vary their use of evidential strategies accordingly. This leads to the proposition of a procedure to quantify and generate features based on the analysis of evidential strategies that can be used to improve corporate bankruptcy prediction. The results presented here stem from an interdisciplinary adaptation of linguistic findings and provide future research with another means of analysis in the area of text mining.

Do key audit matters signal corporate bankruptcy?

Research Question: This paper aims to answer whether the Key Audit Matters (KAMs) Signaling Corporate Bankruptcy and investigate auditor responsibility versus the lack of auditing standards, and examine whether the disclosure of the KAMs by independent auditors enhances the prediction of corporate bankruptcy and the extent to which the KAMs reduce the information asymmetry between firm managers and shareholders. Motivation: We analyse the risk topics in the annual reports, then the KAMs highlighted by the auditor and then the KAMs are disclosed by adopting a case study approach. Data: We use a single descriptive case study approach and read the relative academic and professional literature to explore the KAMs included in the auditors' reports before the Thomas Cook Group Plc bankruptcy. Findings: We find no significant predicting power of KAMs disclosed by Ernst & Young (EY) on Thomas Cook's annual reports. We found that the auditor is not responsible for indicating financial failure. Contribution: We suggest that the regulators and the accounting boards adopt more restrictive standards and improve the International Standards on Auditing (ISA) 701. Furthermore, attention should be focused on the reliability of KAMs specified in ISA 701. We conclude that the KAMs are ineffective in disclosing bankruptcy risk. Our paper concludes that the current auditing standards should be more instructive in preventing corporate bankruptcy. We contribute to the literature in a unique and core research area not researched previously.

Machine Learning for Bankruptcy Prediction in the American Stock Market: Dataset and Benchmarks

Predicting corporate bankruptcy is one of the fundamental tasks in credit risk assessment. In particular, since the 2007/2008 financial crisis, it has become a priority for most financial institutions, practitioners, and academics. The recent advancements in machine learning (ML) enabled the development of several models for bankruptcy prediction. The most challenging aspect of this task is dealing with the class imbalance due to the rarity of bankruptcy events in the real economy. Furthermore, a fair comparison in the literature is difficult to make because bankruptcy datasets are not publicly available and because studies often restrict their datasets to specific economic sectors and markets and/or time periods. In this work, we investigated the design and the application of different ML models to two different tasks related to default events: (a) estimating survival probabilities over time; (b) default prediction using time-series accounting data with different lengths. The entire dataset used for the experiments has been made available to the scientific community for further research and benchmarking purposes. The dataset pertains to 8262 different public companies listed on the American stock market between 1999 and 2018. Finally, in light of the results obtained, we critically discuss the most interesting metrics as proposed benchmarks for future studies.