SARIMA and ANN Approaches in Forecasting the Volume of Postal Services

Abstract

In this paper, time series data forecasting was done by using a seasonal autoregressive integrated moving average (SARIMA) model in XLSTAT add-on for Excel and in MATLAB environment, as well as an artificial neural network (ANN) model. A long short-term memory (LSTM) network was used to construct the ANN model. Both approaches were used for forecasting the volume of express mail services (EMS) in international traffic in the Republic of Serbia and the obtained results were compared with the original data. Significantly better modelling results were obtained by using ANN approach.