ESTIMATING MONEY DEMAND FUNCTION USING CO-INTEGRATION ANALYSIS: THE CASE OF CANADA (1960-2005)

Abstract

This study analyses the demand for money in Canada using a integration econometric analysis for the year 1960 to 2005. The study identifies a number of factors that influence the demand for money in an economy and uses econometric analysis to analyse their influence in the money demand. The question of whether the long-run money demand function can be classified as stable in the long-run, has been a key question for analysis over the years in this field. This paper starts by estimating the long-run money demand model with OLS. The paper then moves on and conducts an augmented Dickey-Fuller unit root tests to determine the order of integration of the variables that affect the money demand in an economy in the long run. Using the estimated residuals from the basic econometric model, we use Engle-Granger procedure to test for co-integration by carrying out an ADF test on the estimated residuals. The specification in levels shows evidence of being a co-integrating relationship since the residuals are stationary and thus meets the Engle-Granger criteria for stationary. Thus, the most obvious result from the regressions carried out is that there is only one co-integrating long-run relationship between the variables included in money demand function. The Keynesian theory of demand for money is borne out by these results, with the transaction demand and precautionary demand having the greatest effect on overall demand for money through income.