Output Gap and GDP in Brazil: a real-time data analysis
Economic agents make decisions using real-time data. However, recent literature has shown that several economic activity measures go through important revisions over time, impairing the reliability of real-time data. We organize a real-time dataset for Brazil’s GDP, and assess the revisions of GDP growth and the output gap. We show that GDP growth revisions are substantial, with a 0.7 p.p. mean absolute revision for the quarter-over-quarter growth, although the revisions become less important for four-quarter changes. To assess output gap revisions, we use four methods to estimate the output gap: Hodrick-Prescott filter, linear trend, quadratic trend, and Harvey-Clark model of unobservable components. The output gap revisions are substantial in all methods, with absolute mean revisions between 0.6 p.p. and 2.3 p.p. In three out of the four methods, the revisions implied changes in the output gap sign in 30 percent or more of the cases. In general, both the GDP data revision and the sample increase are relevant sources of output gap revisions. Key words: Real-time data; Output gap; Gross Domestic Product; Business cycle; Brazil. JEL classification: C82; E32.
When requesting a correction, please mention this item's handle: RePEc:bcb:wpaper:203. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Francisco Marcos Rodrigues Figueiredo)
If references are entirely missing, you can add them using this form.