Forecasting INR/USD Exchange Rates using Hybrid and Neural Network Model

Abstract

Prediction of exchange rates plays a vital role in international trade, stocks and framing the policies of exports and imports. USD exchange rates used widely for many business areas. In this paper an attempt is made to predict INR/USD exchange rates using Hybrid model that joins the forecasts of ARIMA as well as FFNN. The forecasting accuracy of developed models, were tested used the error measures like MAE, MAPE & RMSE. The results shows hybrid model has greater accuracy compared to ARIMA and FFNN model. The predicted exchange rates would vary between 70.80 and 71.39 for the next one month and this variation is exchange rates would help the business people and also for framing the govt policies within the upcoming future.