关键词: 财务危机预警, 粗糙集, 径向基函数神经网络, 主成分分析, 现金流量

Abstract: A new hybrid RPR model for predicting financial distress of Chinese listed corporations was put forth. The proposed model integrates the rough set(RS), the principle component analysis(PCA) method and the radial-basis function neural network(RBFN). Besides the traditional financial indicators, the cash-flow indicators were introduced, which can perfectly reflect the real-time financial situation of a corporation. In the model, the RS was applied to reduce the indicator, the PCA was employed to select indicators and reduce dimensions, the RBFN was used as a prediction tool for the corporate financial situation. Experimental results suggest that the model has high prediction accuracy and execution efficiency.

Key words: rough set, radial basis function network, principal component analysis, cash-flow, financial distress prediction