Abstract:
This study investigates the most significant variables influencing predictive models employed by financial institutions in assessing the likelihood of bankruptcy for small and medium-sized enterprises. The study encompasses a comprehensive analysis, conducted using R software, on a 2018 dataset containing information about various SMEs.
The initial phase of the research involves the testing and examination of predictive models to identify the most significant variables impacting bankruptcy prediction.
Furthermore, the thesis includes a crucial second phase wherein missing data within the dataset are imputed using an algorithm. This process simulates a real-world scenario for banks, where cases are received and evaluated individually. The imputation methodology is designed to enhance the robustness of the predictive models, ensuring their applicability to situations where data may be incomplete.
Through this dual approach, the study aims to contribute valuable insights into the key determinants of SME bankruptcy prediction models, offering practical implications for financial institutions in their risk assessment processes.
