Predicting Financial Distress of Chinese Listed Corporate by a Hybrid PCA-RBFNN Model

Abstract

This paper is to develop a hybrid PCA-RBFNN model for financial distress prediction of Chinese listed corporate. The proposed hybrid model integrates the principle component analysis (PCA) method and the radial-basis function neural network (RBFNN). Besides the traditional finance indicators, we introduce the cash-flow indicators which perfectly reflect the real-time financial situation of a corporate. In our proposed model, the PCA method is employed to select indicators and to reduce dimensions, and the RBFNN is used as a predicting tool for corporate financial situation. The experimental results suggest that the model has high prediction accuracy and execution efficiency.