本研究為國內首篇嘗試應用非線性共變量單根檢定檢測匯率決定理論的貨幣模型之實證研究。我們重新檢驗非線性動態的匯率決定之貨幣模型,從名目匯率與貨幣基要之間的離差來解釋其實證表現,並採用Tsong (2011) 提出的CKSS單根檢定,應用於16個國家在後布雷頓森林時期,探討是否支持長期匯率決定的貨幣模型。本研究使用CKSS單根檢定,實證結果指出:由於捕捉非線性調整以及加入恆定共變量,檢定力大幅提升,除了挪威、瑞士以及義大利這三個國家仍無法找到匯率與貨幣基要間存在共整合外,其他13個國家顯示有足夠的證據支持長期匯率決定之貨幣模型。
This study serves one of the first paper adopting the nonlinear covariate unit root test to examine the performance of monetary models of the exchange rate determination. We reexamine empirical performance of monetary exchange rate with nonlinear dynamics of nominal exchange rate deviation from the monetary fundamentals. We employ CKSS unit root test proposed by Tsong (2011), and test the long-run monetary models of the exchange rate determination for 16 countries using data during the post-Bretton Woods. Our empirical evidence which applies CKSS unit root test shows that owing to capturing of nonlinear adjustment and adding stationary covariates in the regression equation improve the power of unit root tests. In all countries, except Norway, Swizerland and Italy, the nominal exchange rate is cointegrated with the monetary fundamentals. We find support for a simple form of the long-run monetary model in over half of the countries we consider.