A COMPARATIVE STUDY OF GENETIC ALGORITHM AND NEURAL NETWORK MODEL IN BANKRUPTCY PREDICTION OF MANUFACTURING FIRMS IN NIGERIA

Abstract

Previous studies have established the comparative accuracies of statistical failure models in Nigeria. However, the assumptions of these models often limit their practical application. The study, therefore, compares two models developed using AI techniques, the genetic algorithm (GA) and neural network on a sample of quoted manufacturing firms in Nigeria. This study adopts a quantitative approach and utilises a sample of sixty-six (66) companies listed on the Nigerian Stock Exchange (NSE), after excluding firms from the financial, natural, and oil & gas sectors. The study relied on secondary data from annual financial statements. The McNemar test was utilised to compare the accuracies of the two models. The model results showed a significant difference in the classification accuracies of the GA (96.94%; 97.85%) compared with the neural network (92.2%; 94.4%) models. In other words, the GA model outperformed the neural network model in corporate bankruptcy prediction. The inclusion of selected corporate governance variables also improved the accuracy of the models. The results demonstrate the practicality of using GA in a different context from prior western studies with different regulatory and institutional regimes.