6.6 A SAMPLE NEURAL NETWORK PROJECT
• We next describe a typical application of neural networks to predict bankruptcy of companies using the same data and a similar experimental design as used by Wilson and Sharda (1994).
• For comparative purposes, the performance of neural networks is contrasted with logistic regression.
• The Altman(1968) study has been used as the standard of comparison for many bankruptcy classification studies using discriminant analysis and logistic regression; follow-up studies have identified several other attributes to improve prediction performance.
• We use the same financial ratios as in Altman’s study, realizing that more sophisticated inputs to the neural network model should only enhance its performance.