The Importance of Metadata for Analyzing Data Revisions

• Activity Number: 563
• Type: Topic Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Andreas Lorenz*+
• Companies: Deutsche Bundesbank
• Keywords: revision analysis ; real-time data ; metadata ; industrial production

Abstract

Revisions analysis is nowadays widely used for characterizing many dimensions of the quality of economic indicators such as the reliability of first estimates and the size and volatility of later revisions. At the same time, the implications that changes in metadata - ie in definitions and methods underlying official statistics - have for the extent of revisions are seldom explicitly reflected. The paper demonstrates the usefulness of a concise metadata set for analyzing revisions and interpreting the results. First, a revisions analysis of real time data of the German Index of Industrial Production for the period 1995 to 2009 is performed. Second, changes in the revision pattern are traced back to changes in collection and compilation methods and to singular events in the period under investigation. Third, inferences made in the absence and in the presence of metadata are compared. A result is that, without metadata, revisions analysis can be misleading when building expectations on future revisions.