Ain't Behavin': Forecast Errors and Measurement Errors in Early GNP Estimates
- 1 April 1987
- journal article
- research article
- Published by Taylor & Francis in Journal of Business & Economic Statistics
- Vol. 5 (2) , 165-175
- https://doi.org/10.1080/07350015.1987.10509575
Abstract
Revisions of the early GNP estimates may contain elements of measurement errors as well as forecast errors. These types of error behave differently but need to satisfy a common set of criteria for well-behavedness. This article tests these criteria for U.S. GNP revisions. The tests are similar to tests of rationality and are based on the generalized method of moments estimator. The flash, 15-day, and 45-day estimates are found to be ill behaved, but the 75-day estimate satisfies the criteria for well-behavedness.Keywords
This publication has 17 references indexed in Scilit:
- Data Revision, Reconstruction, and Prediction: An Application to Inventory InvestmentThe Review of Economics and Statistics, 1984
- Nearly Efficient Estimation of Time Series Models with Predetermined, but not Exogenous, InstrumentsEconometrica, 1983
- An integrated view of tests of rationality, market efficiency and the short-run neutrality of monetary policyJournal of Monetary Economics, 1983
- Estimation and testing in time-series regression models with heteroscedastic disturbancesJournal of Econometrics, 1982
- Large Sample Properties of Generalized Method of Moments EstimatorsEconometrica, 1982
- Tests of equilibrium macroeconomics using contemporaneous monetary dataJournal of Monetary Economics, 1982
- Flexible Exchange Rates, Prices, and the Role of "News": Lessons from the 1970sJournal of Political Economy, 1981
- A pitfall in estimation of models with rational expectationsJournal of Monetary Economics, 1980
- Money stock revisions and unanticipated money growthJournal of Monetary Economics, 1980
- The Use of Preliminary Data in Econometric ForecastingThe Review of Economics and Statistics, 1978