Testing weak rationality of forecasts with different time horizons
- 1 October 1993
- journal article
- research article
- Published by Wiley in Journal of Forecasting
- Vol. 12 (7) , 541-558
- https://doi.org/10.1002/for.3980120702
Abstract
Starting with conventional tests for weak rationality, we show how additional tests can be performed if predictions with different time horizons are commonly used in them. Next, we show that most of these tests can still be applied if the structure of the economic system changes but only some of them if the true system is unknown. Finally, these tests are applied to the semi‐annual one‐and two‐step predictions of the group of five leading economic research institutes in the Federal Republic of Germany. For the two‐step predictions we find more evidence against the rational expectations hypothesis than for the one‐step predictions.Keywords
This publication has 21 references indexed in Scilit:
- Tests of exchange market efficiency: fragile evidence from cointegration testsJournal of International Money and Finance, 1991
- Measurement Errors and Tests for RationalityJournal of Business & Economic Statistics, 1991
- Unbiasedness, efficiency and the combination of economic forecastsJournal of Forecasting, 1989
- Co-Integration and Error Correction: Representation, Estimation, and TestingEconometrica, 1987
- A Reexamination of the Rationality of the Livingston Price Expectations: NoteJournal of Money, Credit and Banking, 1986
- Formation of Inflation Expectations in British Manufacturing IndustriesThe Economic Journal, 1985
- TESTS FOR RATIONALITY OF THE CARLSON‐PARKIN INFLATION EXPECTATIONS DATA*Oxford Bulletin of Economics and Statistics, 1984
- On Testing for Rationality: Another Look at the Livingston Price Expectations DataJournal of Political Economy, 1978
- Rational Expectations, the Real Rate of Interest, and the Natural Rate of UnemploymentBrookings Papers on Economic Activity, 1973
- The Effect on the t Distribution of Non-Normality in the Sampled PopulationJournal of the Royal Statistical Society Series C: Applied Statistics, 1968