Time and frequency domain tests of some U.S.-Canadian relationships under an autoregressive filter.

Our purpose here is to examine whether this association and causal relationship can be detected using reduced form relationships in the time and frequency domains. We hasten to add that not finding a relationship should not lead one to infer that none exists. The reduced form single equation model will be incapable of detecting a relationship when important other variables are omitted or if one of the variables is a control variable being effectively used to offset the exogenous disturbances on a target variable in one country in such a way as to destroy the correlation between the control variable and the target variable (Kochin, 1972). [ABSTRACT FROM AUTHOR]