Font Size: a A A

Testing international finance equilibrium relationships with a Fourier series to capture structural change

Posted on:2008-07-31Degree:Ph.DType:Thesis
University:The University of AlabamaCandidate:Yuan, JinganFull Text:PDF
GTID:2440390005472473Subject:Economics
Abstract/Summary:PDF Full Text Request
The major goal of this dissertation is to analyze potential nonlinearities in real exchange rate and interest rate behavior in the Pacific Rim nations. Specifically, the paper considers whether econometric models with asymmetric adjustment or with structural breaks best characterize the time-series properties of these variables.;Chapter 1 introduces several important unit root tests, especially the standard Dickey-Fuller test (Dickey & Fuller, 1979, 1981), TAR and MTAR unit root tests, and the Fourier unit root test. The Fourier unit root test differs from previous unit root tests in that it considers structural change without presuming the dates, number, or form of the breaks.;Chapter 2 applies standard Dickey-Fuller unit root test (Dickey & Fuller, 1979, 1981), TAR and MTAR unit root tests (Enders & Granger, 1998) and ESTAR unit root test (Kapetanios, Shin, & Snell, 2003) to study the property of the real exchange rates in Pacific Rim nations from different aspects. The empirical results demonstrate that the TAR-consistent unit root test provides the strongest evidence for PPP, which indicates that the reversion to long-run equilibrium (PPP) has important asymmetric effects.;Chapter 3 employs both the Fourier and standard Dickey-Fuller unit root tests to test the expectations hypothesis of term structure of interest rates in Japan. The empirical results show that, although the two unit root tests both strongly reject the null hypothesis of a unit root, an error-correction model with a Fourier series to capture nonlinearity is preferred over a linear error-correction model when studying interest-rate dynamics.
Keywords/Search Tags:Fourier, Unit root, Test, Structural
PDF Full Text Request
Related items