The Unemployment Rate Forecast Model Basing on BP Neural Network

Unemployment is a major problem in China at the present time, it is also a critical factor relative with the development of society. For this reason, predicting unemployment rate as a method of economic and social studies is being paid more and more attention. An accurate predicted outcome is significant for the government to formulate and promulgate the relative policies. In this paper, according to a set of indicators suggested by experts, a unemployment rate forecast model is constructed. Before the data is inputted into the back-propagation neural network (BPNN), through using principal component analysis (PCA) method to decrease the dimension of the data, it can diminish the scale of BP neural network and then improve the networkpsilas generalization capability. At last, correlation analysis is adopted to check the quality of the training outcome of BP neural network. The result shows, this prediction model has good accuracy and applicability.