Rough Set Neural Network Based Financial Distress Prediction

The training time of the neural network based financial distress prediction is very long when the input volume is large. The paper presents rough set neural network based financial distress prediction method. Through the financial ratios regarded as condition attribute and the enterprise financial status as decision attribute, the decision system of financial distress prediction is constructed. The minimum attribute set is obtained by attribute reduction. The financial ratios in the minimum attribute set are regarded as the inputs of the neural network. The neural network is trained using the training samples and the financial distress prediction model is obtained. The test results show that the training time of the method is shortened obviously and the prediction results are correct and effective.