Abstract
The Purchasing power parity ¡]PPP¡^ theory was originally developed by a Swidish economist, Gustav Cassel, in 1916. It is a method using the long-run equilibrium exchange rate of two currencies to measure the currencies' purchasing power. It is based on the law of one price, the idea that, in an efficient market, identical goods must have only one price internationally. This parity is a central building block of many theoretical and empirical models of exchange rate determination, since most are relied on PPP as the basis for long-run real exchange rates.
While the literature on the PPP hypothesis is voluminous today and still growing, the doctrine has not found well. The validity of PPP can be examined by testing the stationary of real exchange rates. Most of the empirical evidences relied mainly on using linear structure to explore PPP in the past. By using traditional unit root test, the PPP is hard to hold in the long run. There is a growing consensus that previous empirical research reflects the poor power of the tests rather necessarily against PPP. Therefore, the use of more powerful tests is needed.
Recently, an alternative point of view based on the presence of market frictions that impede commodity trade has arisen. The adjustment of real exchange rates is perhaps described more appropriately as a nonlinear process once market frictions are taken into account. There are several reasons that theoretically explain why the adjustment process of deviations from PPP is nonlinear, such as transactions and transportation costs and tariffs and non-tariff barriers to international trade. Therefore, the analysis of real exchange rate should be conducted under the nonlinear structure.
This study uses the STAR methodology proposed by Granger & Teräsvirta (1993) and Teräsvirta (1994) to examine whether the deviation of PPP is a nonlinear dynamic adjustment among the following countries: Australia, Denmark, Italy, Japan, Luxembourg, Norway, Spain, Sweden and the United Kingdom. If the linear hypothesis was rejected, then to distinguish if the model of STAR is LSTAR or ESTAR. This study finds that the deviations from equilibrium exchange rates show strong evidence of nonlinear properties. The deviations of exchange rates for all countries can be explained by the LSTAR model. In conclusion, this study finds the real exchange rates exhibit the property of nonlinear mean reversion for most countries.
| Identifer | oai:union.ndltd.org:NSYSU/oai:NSYSU:etd-0622107-075328 |
| Date | 22 June 2007 |
| Creators | Shen, Hung-Ling |
| Contributors | Ming-Jang Weng, Yung-hsiang Ying, Jai-His Weng |
| Publisher | NSYSU |
| Source Sets | NSYSU Electronic Thesis and Dissertation Archive |
| Language | Cholon |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0622107-075328 |
| Rights | not_available, Copyright information available at source archive |
Page generated in 0.0021 seconds