Modeling and Forecasting of Foreign Exchange Rates Using Auto Regressive Moving Average (ARIMA) and Artificial Neural Networks (ANN)

The exchange rates play a vital role in controlling the dynamics of the exchange market. As a result, the appropriate prediction of exchange rate is a crucial factor for the success of many businesses and fund managers. For more than twenty decades, Box Jenkin's Auto Regressive Integrated Moving Average (ARIMA) technique is one of the most sophisticated extrapolation method for forecasting. It predicts the values in a time series as a linear combination of its own past values, past errors and current and past values of other time series. Artificial Neural Network (ANN) is a modern non linear technique used for prediction that involves learning and pattern recognition. The historical monthly data for the years 1999-2009 (10 years) for five exchange rates namely US Dollar (USD), Great Britain Pound (GBP), Kuwaiti Dinar (KWD), Japanese Yen (JPY), and Hong Kong Dollar (HKD) were modeled using these two techniques and the Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE) Mean Absolute Percentage Error (MAPE) are used to evaluate the accuracy of the models. Results show that ANN model performs much better than the traditional ARIMA model. The main focus of this paper is to forecast the monthly exchange rates using various ARIMA models and ANN models and the future exchange rates is forecasted for the succeeding months.