Comparison of singular spectrum analysis and autoregressive integrated moving-average model on forecasting tourist arrival to langkawi

Abstract

This study investigates the performances of forecasting tourist arrival between Singular Spectrum Analysis (SSA) and Autoregressive Integrated Moving- Average (ARIMA) for the data of tourist arrival to Langkawi by ferry, domestic flight and International flight was used as case study. The forecasting accuracy of each models has been evaluate based on performance of root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). From this study, the result shows that ARIMA model produces minimum values of MAE, RMSE and MAPE for each data of tourist arrival to Langkawi. Besides, the predicted values by using ARIMA model are more accurate with actual values compared to SSA method.