Comparative Studies of Hybrid ARIMA and Artificial Neural Network (ANN) Techniques for Predicting Exchange Rate in Pakistan

For predicting time series data the ARIMA and ANN provides a good technique in the field of research. Time series data often contain both linear and nonlinear patterns. Therefore, neither ARIMA nor neural networks can be adequate in modeling and predicting time series data. When applying the linear models, most existing studies seem to use the same specification for estimation and forecasting, but the dynamic impact of the concerned variables is ignored. In this study combined the ARIMA and Artificial neural network model by adopting both an equal weighted approach and profit weighted approach to capture both th linear and nonlinear components of the exchange rate, and also developed a hybrid techniques by using models of artificial neural network (ANN) and autoregressive integrated moving average (ARIMA) and their performance were compared to ANN and Hybrid ARIMA models. The Hybrid models are used for forecasting the future exchange rate for dollar, the exports and imports of data The findings showed that combining both ARIMA and ANN models reap the advantage of linear and nonlinear modeling. The capability of the two models are analyzed based on standard statistical measures such as, mean absolute error (MAE), root mean square error (RMSE), and mean squared error (MSE). Models effectiveness The effectiveness of the models are analyzed for the foreign exchange rate, imports and exports of the data and concluded that hybrid techniques provided the best forecasting results.