Cointegration and the Aggregate Demand for Money Function
We have argued that in a multivariate context with integrated variables it is important to test for cointegration (i.e., long-run equilibrium relationships). If the variables are integrated of the same order, but not cointegrated, ordinary least squares yields misleading results. In fact, Peter Phillips (1987) formally proves that a regression involving integrated variables is spurious in the absence of cointegration. In this case, the only valid relationship that can exist between the variables is in terms of their first differences.
KeywordsUnit Root Granger Causality Money Demand Error Correction Model Nominal Interest Rate
Unable to display preview. Download preview PDF.