New Application of Computational Intelligence Method of EMD A Case Study of International Tourist Arrivals to China

Empirical Mode Decomposition (EMD), a computational intelligence method dealt with non-linear complex system, was introduced to find out the fluctuation rule of international tourist arrivals in this paper. Main results show that: (1) Through decomposition of long-term and short-term historical data respectively, four IMFs and one residual trend term are both obtained. There are almost the uniform fluctuation periods of the first three IMFs, which are 3, 6, and 12 months separately (Tab.1 and 2). (2) In the long run, the dominant factor to control the change of international tourist arrivals is the residual res whose variance contribution is 83.1%; while in the short term, intense fluctuation of 3 months' period with the biggest variance contribution, 50%, is still the main change characteristic. (3) Intense fluctuation of international tourist arrivals should be paid more careful consideration when establishing recent tourism plan. At the same time, long-term measures to deal with the large tourists flow will also be endeavored in immediately. As one of the best methods of extracting data series, EMD is great beneficial to predict future international tourist arrivals and provide a theoretical guidance for tourism policy.