2006年3月

景氣波動變異對景氣轉折點認定之影響：跨國的實證研究
An Analysis of the Effect of Output Volatility on Turning Points Idetification: International Evidence

研究論文 Research Article

陳仕偉
Shyh-wei Chen

37-76

文獻上有關景氣循環的非線性及共同波動特徵的探討非常多，但有關於景氣波動變異行為的研究則不多，本文以五個先進國家及台灣為實證對象，進行景氣波動變異的跨國研究。我們發現各國的景氣波動的變異都有改變的現象，而且美國、加拿大、英國、義大利、日本及台灣等國所得成長率的變異都有縮小趨勢。實證模型若不考慮這個特點，在認定各國的景氣轉折點的過程中將產生偏誤的狀況，其中以日本及台灣最為嚴重。如何在總體經濟模型內納入景氣波動變異改變的特色，是未來總體經濟學者在建構理論模型時可以思考的一個方向。為了能夠同時刻劃景氣非線性、共同波動行爲及景氣波動的變異行為等三個特徵，我們進一步修正實證模型，並以台灣為例進行探討。

Despite the abundance of studies on the feature of asymmetry and co-movement of the business cycle in the extant literature, rarely has this issue been investigated for the volatility behavior of the business cycle. This paper fills this gap and we investigate the volatility of the rates of output growth for the US, Canada, the UK, Italy, Japan and Taiwan. The empirical results show that there have been structural changes in the volatility of output growth for these countries. While the Markov Switching heteroscedasticity model can capture this feature very well for all six countries, the modified Markov Switching heteroscedasticity model introduced here not only performs extremely well in modeling the volatility behavior of the growth rate of real GDP, but, at the same time, also successfully identifies business cycle peak and trough dates.

景氣波動變異, 非線性, 共同波動, 馬可夫轉換模型
business cycle volatility, nonlinear, co-movement, Markov Switching model

經濟學
Economics

摘要 全文PDF