預測國際旅遊需求的研究最近越來越受到矚目，因此，預測旅遊需求人數是很重要的，至今已經有很多的文獻做過不同國家的旅遊需求分析。近來，中國大陸為世界旅客輸出最大的出境國家，且拜直航之賜，來台觀光團費大幅調降，更是大大提高陸客來台意願，所以，預測大陸來台旅遊的人數對台灣觀光旅遊的規劃與發展是很重要的議題。但開放至目前的時間尚短，相對來說，資料量就略顯稀少，且正因為剛開放，陸客來台旅遊數量未趨穩定，資料變異較大，目前並未有一個適合的預測方法可以應用，故本研究利用灰色預測方法可處理少資料的特性和支援向量機估計風險最低的優點，提出了將兩者混合的預測模式──灰色支援向量迴歸模式(Grey Support Vector Regression, GSVR)；並以MAPE作為預測績效衡量指標。實證結果顯示GSVR模式(MAPE=19.0025%)表現明顯優於簡單迴歸模式(MAPE=39.7775%)、指數平滑模式(MAPE=38.0638%)及GM(1,1)模式(MAPE=46.4863%)，並可成功外插預測2010年1月的大陸來台旅遊人數，MAPE值可達0.96%。 The tourism industry, which benefits the transportation, accommodation, catering, entertainment and retailing sectors, has been blooming in the past few decades. The 20th century witnessed a steady increase in tourism all over the word. Each country wants to know its international visitors and tourism receipts in order to choose an appropriate strategy for its economic well-being. In Taiwan, the visitors from China become possible since 2008 whereas the data is limit, fluctuated and difficult to forecast in tourism planning. In order to cope with this problem, we use the both advantages of support vector regression and grey theory to construct a new model – Grey Support Vector Regression (GSVR). Estimated China tourists were compared with actual published China tourists by MAPE. Empirical results demonstrate GSVR model (MAPE = 19.0025%) showed significantly better than simple regression models (MAPE = 39.7775%), single exponential smoothing model (MAPE = 38.0638%) and GM (1, 1) (MAPE = 46.4863%), and can successfully extrapolation 2010/1 of China tourists, MAPE values up to 0.96%.
