In this chapter we consider and illustrate a solution to the inter-related problems of monetary aggregation and estimation of money demand functions. In doing so, we use quarterly U.S. data and take a demand systems approach. We handle the problem in three stages: (i) we apply revealed preference tests to determine admissible and separable components; (ii) we aggregate these components using a superlative index — the Divisia index; and (iii) we use the basic translog demand system to deal with the problem of money demand.
KeywordsCovariance Income Autocorrelation Dition Volatility
Unable to display preview. Download preview PDF.
- 1.Notice that in the data set available by the Federal Reserve Bank of St. Louis, Super NOW accounts are reported separately from other checkable deposits for the period 1983:1–1985:4 — after 1985, they are included in other checkable deposits. Because the NONPAR tests can only be performed over a constant number of assets, for the period 1983:1–1985:4 we included (using a simple summation) Super NOW accounts in other checkable deposits and calculated the relevant user cost series as a weighted average. We did the same with money market deposit accounts (at commercial banks and thrifts) which for the 1982:4–1991:2 period are reported separately from savings deposits.Google Scholar