本文透過Breitung (2001)提出的非線性秩檢定共整合方法探討人民幣匯率與總體經濟變數的長期均衡關係，分析人民幣匯率與總體經濟變數是否存在長期非線性共整合的效果。實證結果指出，實質匯率、政府支出比率、經濟開放程度、實質貨幣數量及相對生產力皆具有非線性共整合的現象，存在長期均衡關係。我們進一步利用門檻誤差修正模型去探討人民幣匯率與總體經濟變數之因果關係。實證結果發現人民幣匯率與經濟開放程度、政府支出比率、實質貨幣供給量、相對生產力等總體經濟變數，支持以非線性誤差修正的過程，並對中國的各個總體經濟變數提供最佳的詮釋。 This study estimates the long-run equilibrium relationship between Renminbi (RMB) exchange rate and macro-economic variables for China. We apply the non-parametric rank tests proposed by Breitung (2001) to analyze whether RMB exchange rate and macro-economic variables for China exist any nonlinear forms. Empirical results indicate that the degree of openness of the economy, the ratio of government expenditure to GDP, relative productive activity differential, and real money supply that as important explanatory variables of the RMB long-run equilibrium certainly have long run nonlinear relationship. Furthermore, we use the threshold error-correction model (TECM) to detect the nonlinear casual relationship between RMB exchange rate and macro-economic variables are nonlinear forms. In terms of long-run situation, these results demonstrate solid evidence that RMB exchange rate and macro-economic variables for China support the hypothesis of an asymmetrical error-correction process. Chinese macroeconomic variables actually contribute significant effect on RMB exchange rate.
