Prediction And Emperical Study Of RMB Exchange Rate Based On ARIMA-SVM Combined Model

With the rapid development of China's economy and the continuous improvement of its international status,the marketization of the RMB exchange rate has been accelerating,and the RMB exchange rate plays an increasingly important role in the domestic and foreign economies.After the new exchange reform,the uncertainty of the exchange rate trend has been further enhanced,and the exchange rate fluctuations have become larger and larger.Moreover,the gradual deterioration of Sino-US relations has had a major impact on the exchange rate,which has led to an increase in the decline in the RMB exchange rate.In this context,improving the accuracy of exchange rate volatility forecast has important theoretical research significance and practical application value for avoiding foreign exchange risk for micro-subjects such as enterprises and individuals,or for strengthening central bank and relevant administrative departments to strengthen financial supervision and formulate accurate and effective exchange rate policies.Based on the linear and nonlinear characteristics of the exchange rate sequence,this paper mainly analyzes the RMB exchange rate by combining the ARIMA model with the SVM model and taking into account the advantages of the ARIMA model and the SVM model in terms of linearity and nonlinearity.In this paper,the average absolute error(MAE),mean squared error root(RMSE)and mean absolute percentage error(MAPE)are used as the standard to compare the prediction effect of the ARIMA-SVM combined model with a single ARIMA model and SVM model,where MAE and RMSE measure the absolute error between the actual value and the predicted value,and MAPE measures the relative error between the actual value and the predicted value.The results show that the ARIMA-SVM combination model has better prediction effect and higher prediction accuracy than the single ARIMA model and SVM model when predicting the exchange rate of RMB against the US dollar.What's more,the SVM model has higher prediction accuracy than the ARIMA model.