Analysis of Financial Distress Prediction Models for Nigerian Commercial Banks

Accurate financial distress prediction is essential for early warning systems that allow regulators and bank managers to take corrective action before bank failure, and the performance of distress prediction models in the Nigerian banking context deserves professional scrutiny. This study analyses the predictive accuracy of the Altman Z-score, Camels rating framework, and logistic regression models for predicting financial distress in Nigerian commercial banks for the period 2010 to 2022. Secondary data were sourced from the annual reports of fifteen commercial banks, including six that experienced regulatory intervention or distress during the study period. Financial distress was defined as Central Bank of Nigeria-ordered management intervention or capital deficiency. The predictive accuracy of each model was assessed using the area under the receiver operating characteristic curve, sensitivity, and specificity measures. Results showed that the logistic regression model achieved the highest area under the curve (0.87), followed by the Camels framework (0.81) and the Altman Z-score (0.71). The Altman Z-score, developed for manufacturing firms, exhibited lower sensitivity in banking contexts because of differences in asset and liability structures. The logistic regression model identified non-performing loan ratio and liquidity ratio as the strongest early warning indicators. The study concludes that a customised logistic regression model outperforms generic international models for Nigerian bank distress prediction and recommends its adoption by the Central Bank of Nigeria's bank supervision department.

Keywords: financial distress prediction, Altman Z-score, Camels rating, logistic regression, Nigerian banks