Household financial distress in the Visegrad Group countries: comparative analysis using machine learning methods

The article explores financial distress as one of the important aspects of household deprivation, contributing to the discourse on the significance of income-related deprivation. The study aims to examine financial distress and associated household characteristics within the Visegrad Group (V4) countries: Poland, Czechia, Hungary and Slovakia. It applies machine learning methods to analyze the 2022 European Union Statistics on Income and Living Conditions data. The findings reveal that income, education, tenure status, and age are among the key factors influencing household financial hardship in the V4 countries. However, the importance of these factors varies across countries. Additionally, it was observed that CatBoost and LightGBM, as relatively new techniques, outperform more traditional machine learning methods such as Random Forests, Balanced Random Forests, and Extreme Gradient Boosting regarding accuracy and AUC.