Research Of Economic Forecasting Model Based On Artificial Neural Network

With the development of information technology, the change of the rhythm of economic development also is accelerating. How to predict the macroeconomic situation more accurately and reasonably according to the relevant economic indicators has become one increasingly emphasized subject of our government and economic circles. Macro-economy system represents the comprehensive economic status in a country or an area. Finance, as one of the most important parts of national economy, is a crucial macro regulation and control means which is a very complicated system requiring positive and effective regulation. And prediction is needed to be made at first.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 better control or adjustment on the economy. On the basis an efficient prediction or forecast of the future ecnomy. On the basis of predicted result, the decision-makers of government can constitute some plans or projects 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 or accurate or complete, it is difficult to solve such problems by using the traditional prediction methods. Therefore, artificial Neural Networks are applied to prediction.Artificial Neural Network (ANN), booming in recent years, is a system with the function of vast scale and parallel information disposal. It can dispose nonlinear function and relation more easily and efficiently. It has been proved that an ANN of 3 layers with enough nodes can approach nonlinear function with high precision after it is trained fully. So ANN has vast potential in sophisticated and nonlinear modeling. BP(Back Propagation) net is a very important ANN model and used widely. In this paper, I deeply study the economic forecasting methods, classification, prediction accuracy of the evaluation indicators, and artificial neural network internal mechanisms. The BP neural network in artificial neural network is mainly used for the establishment of forecasting model and forecasting GDP which is one of macroeconomic important indicators.