Modeling Money Demand Specification For Canada: An In Depth Analysis
This dissertation builds upon the analysis that I have already undertaken for my Econometrics project titled “Modeling Money Demand in Canada”. It is an attempt to investigate for co-integration by remodeling the Money Demand specification from the project, considering a time span from January 1985 to September 2006: the same as used for the project, using data for Canada.
As a comparison it is also analyzed whether the resulting specification thus uncovered is also valid for the United States of America. This paper is based upon the analysis conducted by some of the key researchers in the field of Money Demand Modeling. Using some of the most recent techniques in Co-integration Analysis, this paper establishes the presence of a long term relationship for the variables considered, both for Canada and the US, thus suggesting a possible specification for the Money Demand function. This dissertation provides a general overview of some of the key issues in Money Demand Modeling and explains various techniques employed in Co-integration analysis.
This paper builds upon the analysis that I have already undertaken for my Econometrics project titled Modeling money demand in Canada. The aim of the project was to model the money demand function (MDF) for Canada, for a specified data set. The real money demand function was specified as M*/P=ƒ (Y, R), using real income (GNP) as the scale variable (Y) and long term interest rate on corporate bonds (R), as the opportunity cost variable. The project was an attempt to check for co-integration between real money demand, real GNP and real interest rates. After initially checking the order of integration of the variables as well as undertaking the relevant diagnostic tests.
Based on numerous theoretical and empirical models, there is evidence of a long term relationship between the variables considered in the model. It is generally perceived that an increase in current income will lead to an increase in the real money demand whereas an increase in the interest rate on bonds, which can be considered substitutes for money, will lead to a decrease in money demanded. The finding of non-cointegration between the variables as suggested by the results of the project; which in essence are the key determinants of the money demand function, might suggest that the model failed to take into account some of the dynamics involved.
The model specified in the project may have been potentially mis-specified, suffering from omitted variable bias or there might have been structural breaks over the data span considered; possibly rendering the estimated money demand function unstable. This may have resulted in the finding of non-cointegration. Another possible reason justifying the lack of a long term relationship between the variables could be the choice of variables used to represent the scale and the opportunity cost variable , also the choice of the monetary aggregate among the competing options, the narrow M1 and the broad M2+ definition of money, can influence the presence of co-integration.
- 20,000 words – 137 pages in length
- Excellent in depth analysis
- Good use of literature
- Excellent use of economic models
- Ideal for economics students
2: Literature Review
3: Data Description
The Order Of Integration
4: The Long Run Equation
5: Co-integration Analysis; Using Engle And Granger Approach
6: VAR Analysis
7: Johansen Approach
8: Impulse Response And Granger Causality
10: Appendix: Canadian Data Set (C)
11: Appendix Section