Back to Search
Start Over
Modeling data revisions: Measurement error and dynamics of 'true' values
- Source :
- Journal of Econometrics, 161(2), 101-109. Elsevier Science
- Publication Year :
- 2011
- Publisher :
- Elsevier BV, 2011.
-
Abstract
- Policy makers must base their decisions on preliminary and partially revised data of varying reliability. Realistic modeling of data revisions is required to guide decision makers in their assessment of current and future conditions. This paper provides a new framework with which to model data revisions.Recent empirical work suggests that measurement errors typically have much more complex dynamics than existing models of data revisions allow. This paper describes a state-space model that allows for richer dynamics in these measurement errors, including the noise, news and spillover effects documented in this literature. We also show how to relax the common assumption that "true" values are observed after a few revisions.The result is a unified and flexible framework that allows for more realistic data revision properties, and allows the use of standard methods for optimal real-time estimation of trends and cycles. We illustrate the application of this framework with real-time data on US real output growth. (C) 2011 Elsevier B.V. All rights reserved.
- Subjects :
- Economics and Econometrics
Observational error
Optimal estimation
Computer science
DATA SET
Applied Mathematics
OUTPUT-GAP
REVISED DATA
GDP
BUSINESS-CYCLE
Data modeling
Data set
Complex dynamics
FINAL VINTAGE
Data revisions
Real-time analysis
REAL-TIME DATA
Econometrics
State space
PROVISIONAL DATA
Real-time data
STATE-SPACE APPROACH
UNITED-KINGDOM
Reliability (statistics)
Subjects
Details
- ISSN :
- 03044076
- Volume :
- 161
- Database :
- OpenAIRE
- Journal :
- Journal of Econometrics
- Accession number :
- edsair.doi.dedup.....0a053bb1741cf5ad81923bd15e1822b8