Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/53902
標題: 維持國際來台旅客人數的時間與空間聚集之計量分析
Temporal and Spatial Aggregation of Sustainable International Tourist Arrivals to Taiwan
作者: 張嘉玲
關鍵字: 應用研究
Tourism and risk management
經濟學
旅遊和風險管理
多變數volatility
預測的風險
空間與時間的聚集
Multivariate volatility
Forecasting VaR
Spatial and temporal aggregation
摘要: 本研究計畫的目的是利用財務經濟學上風險volatility 模型來分析來台國際觀光旅遊人數的成長及風險。有效掌握目前來台國際觀光旅客人數變動的情況及來台旅客最多的幾個國家的成長的風險變動,對於擴展台灣未來的旅遊市場才能有效預測及提出有效之改善政策,因此,研究來台國際觀光旅客人數成長以及其風險變動的評估是有效了解上述問題的重要科學方法之一。如果本研究資料能夠順利取得, 本研究將有二點下列重要貢獻:(1) 文獻上有關觀光旅客的研究都著重在年資料或季資料的分析。無論如何,財務經濟學上利用風險模型 (volatility model) 來分析估計或預測風險的變動時,非常少是拿年資料或季資料分析,大多盡可能的利用月資料、週資料或日資料。基於這些理由,若利用風險模型(volatility model)來評估台旅客成長之風險變動情形,在可能的情況下,應該利用月資料、週資料或日資料是來分析。因此本研究對於時間聚集資料的處理這方面,將是本研究計劃對於目前旅遊經濟研究上新的貢獻。(2) 根據93 年觀光年報,主要來台國際旅遊人數最多的國家分別是日本、港澳、美國及南韓,因此對於來台旅遊客源的分析與不同國家間之空間聚集的問題有密切的關聯。尤其是相對於已經被廣泛利用在旅遊研究上的時間序列分析來說,空間資料的探討至目前為止還未被討論,基於來台觀光旅客客源的聚集問題更應該被謹慎的考慮。因此本研究對於空間聚集資料的處理,是本研究計劃對於目前旅遊經濟研究上另一新的貢獻。此外,本研究還嘗試利用單一變數和多變數模型估計和預測。為了分析旅遊人數及其成長所伴隨的風險預測,本研究模型將嘗延伸模型:預測風險價值門檻(Value-at-Risk (VaR) thresholds)。另外,對於旅遊業稅之風險價值門檻也會盡可能的加以討論。最後, 本研究將希望研究範圍能擴展到來臺旅客相對較少的國家,以作為將來旅遊業和風險管理發展之政策擬定的參考。
The purpose of the project is to model the growth and volatility (or variability inthe growth rate) in daily international tourist arrivals to Taiwan. An analysis ofcurrent tourist arrivals to Taiwan, as well as its associated volatility, is crucial forpurposes of analyzing the rankings of current tourism sources, and the likelihood ofexpanding the market for tourist arrivals in the future. If the required data areaccessible, the research project will lead to two important contributions, as follows:(1) Forecasting tourism arrivals is typically undertaken at relatively low levels ofdata frequency, namely at the annual or quarterly levels. However, volatility is rarelypresent at low frequencies, but can readily be found at higher data frequencies, suchas the monthly, weekly and daily frequencies. For these reasons, it is important toestimate time series models of tourist arrivals and their associated volatility usingmonthly, weekly or daily data, wherever possible. This feature of temporalaggregation is one of the important and novel aspects of the proposed researchproject.(2) As the major international tourism sources for Taiwan are Japan, Macao andHong Kong, USA and the Republic of Korea, the analysis of aggregate tourist arrivalsto Taiwan is inextricably linked to the issue of spatial aggregation across countries.Although the statistical analysis of spatial data is nowhere near as well developed asits time series counterpart, it is crucial for the analysis of aggregate tourist arrivals toTaiwan that spatial issues are considered seriously. This feature of spatial aggregationis another important and novel aspect of the proposed research project.Both univariate and multivariate models will be used to compare the estimatesand forecasts. The models and analysis will be extended to forecast Value-at-Risk(VaR) thresholds for purposes of analyzing the risks associated with the volatility intourist arrivals and their associated growth rates. The possibilities of introducing atourism tax associated with forecasts of VaR thresholds will be entertained. Finally,the project will analyze the possibilities of extending the analysis of Taiwan to othercountries where tourist arrivals are considerably lower than they might otherwise be,in order to develop a marketing policy for tourism and risk management.
URI: http://hdl.handle.net/11455/53902
其他識別: NSC97-2410-H005-004
文章連結: http://grbsearch.stpi.narl.org.tw/GRB/result.jsp?id=1659421&plan_no=NSC97-2410-H005-004&plan_year=97&projkey=PF9709-0168&target=plan&highStr=*&check=0&pnchDesc=%E7%B6%AD%E6%8C%81%E5%9C%8B%E9%9A%9B%E4%BE%86%E5%8F%B0%E6%97%85%E5%AE%A2%E4%BA%BA%E6%95%B8%E7%9A%84%E6%99%82%E9%96%93%E8%88%87%E7%A9%BA%E9%96%93%E8%81%9A%E9%9B%86%E4%B9%8B%E8%A8%88%E9%87%8F%E5%88%86%E6%9E%90
Appears in Collections:應用經濟學系

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.