DECOMPOSITION OF ECONOMIC RELATIONSHIPS BY TIMESCALE USING WAVELETS -MONEY AND INCOME

Economists have long known that timescale matters in that the structur e of decisions as to the relevant time horizon, degree of time aggrega tion, strength of relationship, and even the relevant variables differ by timescale. Unfortunately, until recently it was difficult to decom pose economic time series into orthogonal timescale components except for the shea or long run in which the former is dominated by noise. Wa velets are used to produce an orthogonal decomposition of some economi c variables by timescale over six different timescales. The relationsh ip of interest is that between money and income, i.e., velocity. We co nfirm that timescale decomposition is very important for analyzing eco nomic relationships. The analysis indicates the importance of recogniz ing variations in phase between variables when investigating the relat ionships between them and throws considerable light on the conflicting results that have been obtained in the literature using Granger causa lity tests.