Abstract

Financial risk early warning model for quoted company using data mining technology

Author(s): Wang Zhu

This paper aims to put forward a financial risk early warning model for quoted company, which is an important for modern company management. Firstly, the index system for financial risk early warning is given, which includes 1) Macro-economy, 2) Governmental capabilities, 3) Foreign trade, 4) Currency risk, 5) Financial institutions liquidity risk, and 6) Financial institutions operational risk. Secondly, financial risk early warning model based on radial basis function neural network is illustrated. For the radial basis function neural network, widths and centroids should be fixed, and then weights are obtained through solving a linear equation. Particularly, in our proposed model, nearest neighbor clustering algorithm is exploited to choose the centroids of clusters. Finally, experiments are conducted to make performance evaluation. Compared with other methods, we can see that the proposed is effective to tackle the problem of financial risk early warning for quoted company.