CMStatistics 2018: Start Registration
View Submission - CFE
Title: Measuring GDP growth data uncertainty Authors:  Ana Galvao - University of Warwick (United Kingdom) [presenting]
James Mitchell - University of Warwick (United Kingdom)
Abstract: Economic statistics are prone to data uncertainty since they are subject to both sampling and non-sampling errors. GDP estimates are regularly revised as new information is received and methodological improvements are made. Although the Office for National Statistics, in the UK, emphasise that initial estimates of GDP values will be revised, it is the Bank of England that provide quantitative estimates of the likely uncertainty around past GDP values. We find that the revision mean and measurement error volatility are time varying for both UK and US GDP growth. We evaluate the Bank of England's predictive densities for revised GDP growth values; and show that their point predictions better anticipate mature ONS growth estimates than the ONS's own first releases. Their density estimates are on average well-calibrated, but this masks changes in predictive performance. We propose a measure of data uncertainty that removes the forecastable component of data revisions as predicted by the Bank of England's backcast densities. We show that data uncertainty jumps at the onset of the 2008/2009 recession and contribute to macroeconomic uncertainty.