領先指標對台灣景氣趨勢預測能力的評估

無論是一般民眾的消費、儲蓄計畫，或是廠商的投資決策和政府政策的規劃與施行，均深受景氣變動的影響，因此準確地預測台灣未來的景氣動向，以及認定景氣循環的轉折點、景氣擴張期、收縮期，已成為經濟學者、政府機關和一般大眾所關注的焦點。而早在1930年左右，美國國家經濟研究局(National Bureau of Economic Research, NBER)就已經建立了一套標準與程序，編製各種景氣指標，用以認定美國的景氣循環轉折點的日期，經濟專家學者也根據NBER所編製的景氣指標，建構較嚴謹的經濟理論和計量方法，希望能夠準確地認定景氣的轉折點、循環的來源和波動發生的背景因素。

Business cycle fluctuations have long been a major concern of the government and the investor community. The Council for Economic Planning and Development of the Executive Yuan is responsible for compiling the Business Cycle Index (including leading and coincidental indicators) and the Business Monitoring Indictors, which provide important sources of information to both government policymakers and the investing public. The accuracy of these indexes in predicting changes in business conditions is, understandably, of great interest to economic scholars, government agencies, and the public at large. The purpose of this article is to assess the ability of the macroeconomic and financial variables contained in these indexes to accurately forecast forthcoming changes in Taiwan’s business climate.
Using the traditional linear Granger causality test to empirically examine the relationship between variables in the index of leading indicators and changes in Taiwan’s business conditions, we find that a casual linear relationship does exist between the two. However, earlier literature points to the simultaneous existence of an asymmetric or non-linear causal relationship at work in business cycle fluctuations that the linear method is unable to detect. Our study uses the non-linear causality test proposed by Hiemstra and Jones to investigate and resolve this problem. One of our empirical findings is that there is a non-linear causal relationship between some of the variables that comprise the index of leading indicators. An important policy implication is that decision makers must not overlook the significance of information representing this non-linear influence when evaluating forecasts of future busin es s trends.