Comparative analysis of hybrid models for prediction of tourist arrivals

Abstract

Accurate information on future tourist arrivals is a fundamental key for tourism planning and management. Traditionally, single models have been introduced to predict the future value of tourist arrivals. However, single models may not be suitable to capture the nonlinear and non-stationary nature of the data. In this study, combination method based on Empirical Mode Decomposition (EMD), wavelet and Support Vector Machine (SVM) model, referred to as EMD-WSVM is introduced. This study also presents comparison between the proposed model of EMD-WSVM with hybrid Empirical Mode Decomposition and seasonal autoregressive integrated moving average (EMD-SARIMA) and wavelet with support vector machine (WSVM) model proposed by previous researchers. These models are ranked based on three statistical measures namely Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and correlation value. The results show that EMD-WSVM ranked first based on measures for Cambodia tourist arrivals. The study concludes by recommending the application of an EMD-based combined model particularly with wavelet method reduction approach for tourist arrivals forecasting due to better prediction results.