Innovative Approaches to Evaluating Credit Risk: Data, Models and Strategies

Dear Colleagues,

This Special Issue investigates corporate default modeling, with particular reference to advanced methodologies and alternative data. The ability to predict corporate failure has been extensively explored in the modern credit risk literature. This topic has consistently garnered significant attention, especially when considering small- and medium-sized enterprises, given their importance in the EU economy. Today, this interest is even more pronounced thanks to the introduction of novel statistical algorithms that excel in forecasting bankruptcy.

Machine Learning (ML) algorithms are gradually finding their way into various aspects of credit risk management, including credit scoring and monitoring, due to their remarkable predictive capabilities. Simultaneously, the adoption of these new algorithms is reshaping the focus of modeling. While their performance surpasses that of traditional linear models, there is a growing emphasis on the interpretability of the rules they generate.  Indeed, the scientific community has concentrated on white-box techniques that create explicit models that are interpretable. In line with the guidelines set forth by the European Commission, predictions, no matter how accurate they may be, must be accompanied by explainability measures. Accordingly, the application of post-hoc methods is generating new areas of research on black-box models interpretation.

Using alternative data or the combination of alternative data, covering different aspects of a firm’s workings, also contributes to a more complete picture of the phenomenon of bankruptcy prediction. Most approaches have historically focused on financial and accounting data, sometimes supplemented with macroeconomic factors like interest rates and stock market performance. However, a contemporary trend is shifting toward the integration of diverse datasets. A smaller subset of studies has embraced softer information, including qualitative insights related to management, corporate governance details, or relational data, along with information about the relationship between banks and firms. Additionally, some contributions have extended their scope to encompass contextual information, such as the local banking landscape and the broader institutional framework within which SMEs operate. Incorporating management earnings also contributes to enhanced predictive accuracy. Recent studies have taken a step further by incorporating a firm's historical data into their models, demonstrating that the overall trajectory of these variables can significantly influence the likelihood of default. Moreover, the longitudinal aspect has been extended to include past instances of minor defaults, revealing that many firms that experience such setbacks are indeed able to persevere and continue their operations in the market. Cutting-edge research is now exploring the predictive potential of textual analysis, introducing text-based attributes linked to the creditworthiness of borrowers. Additionally, novel financial sentiment metrics were developed through the analysis of textual reports and web-scraped indicators were built to offer fresh insights into the default prediction landscape.

The dataset complexity in the credit risk field represents a challenging topic. It is very common to encounter data-related problems such as data imbalance, outliers, noisy data, etc. As it turns out, dataset complexity is a major deterioration factor for prediction models. Accordingly, new solutions to overcome the problems related to data complexity are required.

Contributions dealing with original applications and solutions in the field of corporate default assessment are very much welcome.

Dr. Caterina Liberati
Dr. Lisa Crosato
Dr. David Veganzones
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Risk and Financial Management is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• machine learning solutions, including ensemble models
• variable selection
• explainable artificial intelligence
• interpretation of black-box models, including white-box techniques
• resampling techniques
• cross section and longitudinal studies
• cross-countries comparisons
• spatial modeling
• alternative data/big data/open data and integration of alternative sources
• resilience of bankruptcy prediction models to crises