Recalculation of the first Hungarian bankruptcy-prediction model using neural networks
Languages of publication
The article attempts to say whether the latest bankruptcy prediction techniques are more reliable in Hungary's case than traditional mathematical/statistical ones. Simulation experiments carried out on the database of the first domestic bankruptcy-prediction model show clearly that neural-network models possess greater classification accuracy than the models based on discriminant analysis and logistic regression analysis elaborated in the 1990s. The article presents the main results, analyses the reasons for differences, and draws up constructive proposals for further development of domestic bankruptcy-prediction practice.
Publication order reference
CEJSH db identifier