Data Revisions and ISO Vintages

• 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): Jorge Dias*+ and Magdalena Lalik
• Companies: European Central Bank and European Central Bank
• Keywords: data revisions ; real-time data ; ragged edge ; balance of payments

Abstract

Collecting statistical data is a repetitive process that takes time to feed into published time series. Typically, statistical time series are released with a time lag and are subject to revisions that are caused by gradual inclusion of additional information as it becomes available. Consequently, the information content of recent observations within a time series is not uniform, creating the problem of a comparability of observations. This issue, which has been identified in the literature as the "ragged edge" problem, is amplified when several time series are modelled simultaneously. This paper presents an approach for the harmonisation of the information content of time series observations. Real-time databases and patterns of revisions to time series are used to create "ISO vintages" - i.e. time series containing observations with similar maturities. This model is generally available to any researcher, and can be used efficiently in large-scale applications by statistical agencies. The model has been applied to euro area balance of payments statistics and shows that the method produces reliable and efficient estimates for the most recent observations.