The Fisher hypothesis: examining the Canadian experience

In this study the validity of the Fisher hypothesis is investigated for Canada under both fixed and floating exchange rate regimes. An empirically distinction is drawn between the weak and strong form of the Fisher hypothesis. The Johansen-Juselius (JJ) multivariate cointegration methodology is applied to test the weak form while the Phillips–Hansen fully modified ordinary least squares (FM-OLS) technique is used for the strong form hypothesis. The caninical correlations (JJ) methodology has the smallest bias and dispersion and hence is the best among the alternative testing procedures available. The FM-OLS procedure, on the other hand, allows for an unrestricted cointegration test correcting for both endogeneity in the data and asymptotic bias in the coefficient estimates. The Fisher hypothesis is soundly rejected.