This research explores the prediction of bank bankruptcy in Ukraine using logistic regression, a statistical method specifically designed for binary outcome. A comprehensive set of 22 financial indicators, encompassing liquidity, solvency, profitability, and business activity ratios, were extracted from the National Bank of Ukraine to construct the predictive model. Rigorous analysis revealed that among the multitude of indicators, only two emerged as significant predictors of bank bankruptcy: administrative and other operating expenses relative to assets, and interest income relative to assets. These findings underscore the importance of operational efficiency and interest income generation in maintaining bank stability. The logistic regression model demonstrated robust predictive capabilities, accurately classifying 88.9% of banks as either bankrupt or stable. The results of this study highlight the critical need for early warning systems to identify banks at risk of failure. By employing advanced statistical modeling techniques, financial regulators can proactively implement measures to mitigate systemic risks and safeguard the stability of the banking sector, particularly in economies like Ukraine that are susceptible to economic shocks.