A Modified Adaptive Genetic BP Neural Network with Application to Financial Distress Analysis

Neural networks (NNs) have been widely used to predict financial distress because of their excellent performances of treating non-linear data with self-learning capability. However, common neural networks often suffer from long convergent processes and occasionally involve in a local optimal solution that more or less limited their applications in practice. To overcome the drawbacks of neural networks, this study develops a new modified genetic algorithm-multi-population adaptive genetic back-propagation algorithm (MAGBPA). In this paper, a hybrid system combining feed-forward neural network and MAGBPA-multi-population adaptive genetic back-propagation neural network (MAGBPNN) is proposed to overcome NN´s drawbacks. Furthermore, the new model has been applied to financial distress analysis based on the data collected from a set of Chinese listed corporations, and the results indicate that the performance of MAGBPNN model is much better than the ones of common neural network model.