全球金融海嘯後，社會大眾對於景氣的變動更加關注，皆希望能推估和預測景氣的走勢。然而，過去學者在探討景氣指標相關議題時，大多將各景氣構成項目拆開成一個個獨立項目，探討各個構成項目是否有其代表性，或者嘗試建構新的景氣指標構成項目。本研究則使用序列型樣探勘中的SPADE演算法，可以一次綜合多個指標，同時討論景氣領先指標、進出口值與景氣落後指標的先後關聯與趨勢。 研究結果針對海關出口值的部份，選出序列型樣共24項，包含的景氣指標可歸納整理成五個，分別為外銷訂單指數、製造業存貨率、工業及服務業經常性受僱員工人數、SEMI半導體接單出貨比、股價指數。針對機械及電機設備進口值的部份，選出序列型樣共3項，皆與金融業隔夜拆款利率有關。本研究篩選出有關聯的序列規則，亦針對各項序列規則做整理歸納，並與其他相關論文研究的結果以及國家發展委員會的編製進行比較，嘗試佐以理論和海運相關知識，給予序列規則一個合理解釋。;After the global financial tsunami, the public is more concerned about the changes in the economy and willing to foresee economic trends. From past research, the business indicator was separated into component series to explore its correlations and individual representation, or trying to structure a new component series. In this study, we integrated multiple component series while discussing business indicators. For the export, we selected a total of 24 sequential patterns categorized into five groups, relevant to the Index of export orders, Inventories to sales ratio for manufacturing, Regular employees on payrolls in industry & services, SEMI book-to-bill ratio, and TAIEX average closing price. For the import, we selected 3 sequential patterns related with Interbank overnight call-loan rate. In this study, we discover the sequential associations between business indicators and customs export import indicators. The findings are summarized and compared with relevant research result and component indicators frame made by National development council. We also try to comment about it through theory and shipping industry knowledge.
