Predicting corporate bankruptcy using quantitative and qualitative data: A comparative analysis using logit and artificial neural networks

The purpose of this study is to compare the predictive strength of a logistic regression model to two artificial neural network in determining corporate financial distress (bankruptcy). This research expands existing bankruptcy models on two levels. First, it utilizes financial ratios based on the cash flow statements in addition to the usual ratios based on balance sheets and income statements. It then incorporates qualitative data that may indicate financial distress such as change in management and change in auditor. Results indicate that the inclusion of qualitative data enhance the predictive power of both the stastical and neural network models.