A New Decomposition Ensemble Approach for Tourism Demand Forecasting: Evidence from Major Source Countries

摘要: The Asian-pacific region is the major international tourism demand market in world, and its deeply affected by various factors. Previous studies have shown that different factors influence at timescales. Accordingly, decomposition ensemble learning approach proposed to analyze impact of on demand, potential advantages method forecasting Asia-pacific are further explored. This study carefully explores multi-scale relationship between tourist destinations source countries, decomposing corresponding monthly arrivals with noise-assisted multivariate empirical mode decomposition. With China Malaysia as case studies, their respective results show significantly better than benchmarks which include statistical model, machine deep terms level accuracy directional accuracy.