Forecasting short-term tourism demand with a decomposition-ensemble strategy

Tourism demand forecasting has very important reference value for all stakeholders including government policy makers, scenic area managers and tourists. However, most of the current researches only focus on the medium and long-term tourism demand forecasting, and pay less attention to the short-term demand fluctuations. Considering that short-term tourism demand usually has the characteristics of high volatility, nonlinearity, and unstability, this paper introduces the decomposition-ensemble strategy and the classic machine learning algorithm at the same time, and proposes the SSA-SVR model to predict the daily passenger flow volume data of the Mt. Siguniang Scenic Area, a desired result has been achieved.