GDP 是建構景氣基準循環時一個最主要的參考依據。由於 GDP 數列本身為季資料,而基準循環則常需以月資料的型態公佈,因此在建構基準循環時必需要將季資料轉換成月資料以符合計算上的需求。目前經建會的處理方法為利用平均分配的方式來轉換資料,但這樣的轉換公式太過簡化而失去其原有的意義。本研究主要目的是依據 state-space 計量模型,嘗試重新建構適當的GDP 月資料指標以供政府經濟決策之參考。從計量經濟的角度來看,此一模型的優點在於其能包含文獻上常用的其它模型,並且放寬即有共整合的假設,以建構出適當的GDP 月資料指標。若從實證結果來看,本研究所估計出的 GDP 月指標與 GDP 季資料有許多相似的特徵。故我們相信,利用此模型所建構出的月指標確實可以提供政府短期經濟決策時之參考。
In this project, we describe a framework that nests a great variety of interpolation setups and relaxes the co-integration conditions of temporal disaggregation in the literature. Our goal is to evaluate alternative interpolation models and then to produce a monthly deseasonalized Taiwan's real gross domestic product and make it available for researchers and practitioners.
Our empirical result shows that the monthly estimates, incorporated with the information obtained from the related series, are consistent with the quarterly figures. These estimates could be very helpful for short-run policy analysis by signalling any emerging economic problems.
|