A Study On The Predictability Of Guangdong Province's GDP Based On ANN Methods

GDP (Gross Domestic Product) is one of the most important factors in measuring the whole economic situation and status of one country or district. In order to implement a better control or adjustment on the macro-economy, we should firstly make an efficient prediction or forecast of the future economy. On the basis of the predicted result, the decision-makers of government can constitute some plan or project to restrain or stimulate the economy growth. In the current methods, the most common statistics methods are time-series and regression predictions. And macro-economy is a nonlinear system, which keeps changing. Besides, additional interference factors have direct effects on the operation of macro-economy systems, greatly influencing the prediction results. Since the historic data needed for macro-economy models are not stable, not accurate and not complete, it is necessary to solve such problems by using the traditional prediction methods. Therefore, artificial Neural Networks are applied to prediction.ANN(Artificial Neural Network) is an rising cross subject, and in the recent years, it has been used in more and more sphere in society, industry and science, which display a great application foreground in these field. Consequently, the thesis adopt ANN method to predict the GDP of Guangdong province. First of all, All economic data is changed into region between -1 and 1, and then sent into BP neural network receiving training. Parameters derived from the network can be used to forecast and test the economic index. The result can be used to make a comparison with the one of traditional time series methods. Finally, the thesis make a conclusion, the effect of ANN predicting excelled the one of traditional time-series method.