National Statistics Online - Articles - Revision analysis of initial estimates of key economic indicators and GDP components
Articles Database
Revision analysis of initial estimates of key economic indicators and GDP components
The results of revisions analysis work by the Office for National Statistics, this article updates previous articles, sets out challenges and asks questions about readers’ use of the results.
This article sets out the results of revisions analysis work by the Office for National Statistics (ONS), updating the articles published in April and August 1996; sets out some current challenges and asks questions about readers’ use of the results. Readers’ comments are welcome.
The first article in this series was published in the May 1992 issue of Economic Trends. That article reported the results of the analysis of revisions to initial estimates of headline ONS official statistics, based on 10 years of data up to the end of 1991. It was entitled ‘Testing for bias in initial estimates of key economic indicators’. The set of economic indicators analysed were those specified in the CSO’s Agency Framework Document (henceforth referred to as ‘the Framework Document’) in November 1991. Since then the analysis has been successively updated and extended.
As in the previous articles, we analyse revisions to initial estimates of key indicators as defined in the Framework Document. The key economic indicators are all defined as percentage growth rates or percentages of gross domestic product (GDP) or of some other appropriate denominator, so that they are comparable in all periods, regardless of changes in the size of the economy. The following definition of bias is used: revisions are considered to be biased if the mean revision is significantly different from zero.
Download
