A neural network based regional economy early warning system

Regional economy early warning system theory is using a set of supervising, assessing and forecasting index to calculate the tendency of regional economic market on the basis of its cycle research. Contrasted with the traditional early warning theories, the neutral network owns advantages on optimization, function approaching; model reorganization, self-study and self-adaptation. In this paper, an early warning model, which combines BP neutral network with principal component analysis (PCA), is presented to enhance the exact rate of early warning judgment on the economic distress. At last, the model is used to carry early warning analysis on hangmen, Guangdong, China regional economy. The favor results show that BP neural network is very applicable to the foundation of regional economy early warning system.