Research Of Exchange Rate's Prediction Method Based On Neural Network

With the explosion of modern science and technology as well as the development of socialized production and the degree of internationalization of the substantial increase, integration of the world economy is one of the basic characteristics of social development. Financial internationalization since the 1970s and 1980s is the latest dramatic development, financial infiltration profound and wide-ranging is more prominent than any time in the past. And the exchange rate as an important link of international financial relations has been infiltrated into almost all areas of economic life, and play an increasingly important role. Since China's accession to the WTO, financial openness constantly increasing, the corresponding exchange rate risks increasing all the time. In particular, the future of China's reform, the risk of exchange rate fluctuations over the past greatly. Therefore, the correct analysis and forecast for the fluctuations in the exchange rate is undoubtedly of great significance for monetary policy and investment decision-making.The article first focuses on artificial neural network algorithm, as well as BP, then a detailed analysis of a variety of commonly used single forecasting methods and several linear combination of forecasting methods. Among the many individual prediction method, selected two more suited to the exchange rate forecasts predict the combination of nonlinear methods forecast. Combining artificial neural network technology, the establishment of the exchange rate of non-linear prediction model based on a neural network, and the model applied to the real exchange rate forecast. Forecast results show that the method is superior to single forecasting methods and linear combination of forecasting methods, can effectively carry out exchange rate forecast.