Directions of Development and Research Gaps in the Prediction of Bank Bankruptcy or Restructuring

Abstract

Theoretical background: The analysis presented in the article is grounded in contemporary theories concerning the prediction of banking crises and the application of predictive models in the financial sector. Particular emphasis is placed on concepts related to big data, artificial intelligence, and machine learning, which are increasingly pivotal in financial stability analyses. The article references global models such as logit and random forests, highlighting their potential and limitations within the specific context of local conditions, particularly in the Polish banking market. Key gaps in the existing literature are identified, underscoring the necessity of developing models that incorporate macroeconomic variables and local regulatory frameworks, thereby enabling the adaptation of global tools to regional environments more effectively.

Purpose of the article: The article aims to analyze the emerging trends in bankruptcy and restructuring prediction models for banks and identify key research gaps in the context of the Polish banking market, comparing it to global benchmarks. It discusses the increasing role of big data, artificial intelligence, and machine learning in predicting banking crises and assesses their applicability to smaller economies like Poland. The study examines global prediction models, such as logit, random forests, and big data-driven approaches, emphasizing the need to adapt them to local economic conditions.

Research methods: The study employs a comprehensive literature review methodology, critically evaluating existing predictive models and their applicability to the Polish banking market. A comparative analysis is conducted between global predictive models and local market conditions, with a focus on macroeconomic variables such as inflation, unemployment, and exchange rate volatility. The research also examines the challenges specific to Poland’s banking sector, particularly cooperative banks, and explores the integration of advanced technologies, such as big data and artificial intelligence, to enhance predictive accuracy. The methodological approach encompasses critical analysis, document review, and comparative evaluations, supported by expert insights, to identify key research gaps and propose tailored solutions for smaller economies like Poland.

Main findings: The study identifies significant gaps in the application of global predictive models to the Polish banking sector, particularly with regard to the unique challenges posed by local economic conditions and regulatory frameworks. It highlights the necessity of developing tailored models that integrate macroeconomic variables specific to Poland, such as inflation and currency volatility, to improve the accuracy of predictions. The findings underscore the growing potential of advanced technologies, including big data and artificial intelligence, in enhancing early warning systems for banking crises. Additionally, the research provides a comparative perspective, demonstrating the need for greater alignment between global approaches and local market realities, especially for cooperative banks and smaller financial institutions operating within the Polish economy.