Seasonal and trend forecasting of tourist arrivals: An adaptive multiscale ensemble learning approach

In this study, we propose an adaptive multiscale ensemble (AME) learning approach, which consists of variational mode decomposition (VMD) and least square support vector regression (LSSVR) for seasonal trend forecasting tourist arrivals. the formulation AME original arrival series is decomposed into trend, seasonal, remaining volatility components. Then, ARIMA used to forecast component, SARIMA LSSVR The empirical results demonstrate that our proposed approach can achieve higher accuracy.