The conduct of time series analysis on the Euro Area currently presents problems in terms of availability of sufficiently long data sets. The ECB has provided a dataset of quarterly data from 1970 covering many data series in its Area Wide Model (AWM), but not for a number of important financial market series. This paper discusses methods for producing such backdata and in the resulting difficulties in selecting aggregation methods. Simple applicaiton of the AWM weights results in orders of magnitude difference in financial series. The use of different aggregation methods across series induces relationships. The effects of different possible methods of constructing data are shown through estimation of simple Taylor rules, which result in different weights on output gaps and inflation deviation for what are purportedly the same data
The text is part of a series 2006 MMF Conference Papers Number 99
Classification:
C82 - Methodology for Collecting, Estimating, and Organizing Macroeconomic Data ; C43 - Index Numbers and Aggregation ; E58 - Central Banks and Their Policies