EFFICIENT FORECASTS OR MEASUREMENT ERRORS? SOME EVIDENCE FOR REVISIONS TO UNITED KINGDOM GDP GROWTH RATES*

Previous studies of revisions to key economic time series for the United States have suggested that, in some cases, the revisi ons are like pure measurement errors and, in others, that they are not forecastable from information available at the time the data is compiled. In this study, the authors analyze these issues for an extensive set of revisions to estimates of GDP for the United Kingdom and draw out some of the implications of their findings for statistical practice. Copyright 1992 by Blackwell Publishers Ltd and The Victoria University of Manchester