Research of RMB Exchange Rate Forecasting based on NARX Model with NDF

摘要

Nonlinear Auto Regressive models with exogenous Inputs which we use NARX for short,is a branch of Artificial Neural Network.As the forecasting of exchange rate is a nonlinear problem,in this paper,we use NARX model to forecast the RMB exchange rate.To NARX models,selection of input variables has significant influence on the accuracy of the forecasting results.This paper selects non-deliverable forward exchange rate,which is known as NDF,as an external input,use data before and after the reform of the RMB exchange rate during to do the forecasting,and achieved good results.