Autore
Ehrbeck, Tilman

Titolo
Rejecting rational expectations in panel data : some new evidence
Periodico
European University Institute of Badia Fiesolana (Fi). Department of Economics - Working papers
Anno: 1992 - Fascicolo: 91 - Pagina iniziale: 1 - Pagina finale: 44

Early empirical tests for rationality In survey expectation data have either used summary data and thus neglected the resulting aggregation bias or used individual data but ignored the likely correlation of forecasts due to the same aggregate shock surprising economic agents. Two recent papers took these problems Into account and arrived at exactly opposite conclusions: One rejected the Rational Expectation Hypothesis, the other did not. This paper adds evidence from a new data set which features three advantages none of the previously used data sets could combine: (a) Names of forecasters are given along with the forecasts so that they have an incentive to do as well as possible; (b) forecasters predict a quoted price so that there is no ambiguity as to what they are trying to forecast; (c) forecasts for the same target period are made at different points in time so that alternative implications of the Rational Expectation Hypothesis can be tested. Three different tests are employed In this paper. They all reject the Rational Expectation Hypothesis.



Testo completo: http://hdl.handle.net/1814/405

Esportazione dati in Refworks (solo per utenti abilitati)