Abstract:

Forecasting is very helpful tool for making decisions and plans for upcoming time periods. In this study Seasonal Auto Regressive Integrated Moving Average (SARIMA) and Artificial Neural Networks (ANN) models are used to forecast the Pakistan’s import prices of black tea by using data for the time period Jan 2004 to Dec 2014. The performance of SARIMA and ANN models are evaluated on the basis of root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE). The selected model under Box-Jenkins is SARIMA (0, 1, 1)*(0, 1, 1)12. ANN models with different combination of input, hidden and output layers were tested with four activation functions (semi linear, sigmoid, bipolar sigmoid and the hyperbolic tangent function). This study reveals that ANN model performed well as compare to SARIMA model because RMSE, MAE and MSE of ANN model are less as compare to RMSE, MAE and MSE of SARIMA model. Therefore, ANN can be effective for forecasting the Pakistan’s import prices of black tea. Key Words: SARIMA, ANN, Black Tea, RMSE, MSE, MAE.

Keywords: