| Summary: | 碩士 === 淡江大學 === 財務金融學系碩士在職專班 === 93 === Since the era of floating exchange rate in 1970s, there’re so much researches about international finance. According to the different macro-economic setting and based on the different assumption about sticky and flexible price in the short and long term, scholars research exchange rate determination models and develop various monetary models. There are two different conclusions between the flexible-price monetary model (Chicago school) and the sticky-price monetary model (Keynesian school) about the relationship between interest rate differential and exchange rate. In generally, there is no strong evident to totally support any one theory about this relationship.Existing structural models have little in their favor beyond theoretical coherence. Positive results, when they are found, are often either fragile, or unconvincing in that they rely on implausible theoretical or empirical models”. The Panel threshold model is different from other traditional linear econometric mode. This research is the first one to use Panel threshold model as the econometric model to fit and estimate the non-linear relationship in the generalized-sticky price model, modified model by scholars. This research tests the asymmetry adjustment between interest rate differential and exchange rate, if there exists any non-linear relation panel threshold effect between interest rate differential and exchange rate. It tries to resolve the puzzle of traditional linear model to enhance the predictability of monetary model.Using the data of 13 countries in OECD vs. U.S. as sample, this research tests the panel threshold effect between interest rate differential and exchange rate. This research tries to explain the movement of exchange rate affected by the change of interest rate differential between countries and U.S. According to the result of the single panel threshold model, in the period of investigation, the threshold split interest rate differential into two parts, and the connection between exchange rate and interest rate differential are positive and negative correlation, respectively. The empirical evidence shows that there may not exist threshold effect between interest rate differential and exchange rate. The result of the panel threshold effect is not significant.
|
|---|
