Re-examining Rational Expectations
Forthcoming: Developments in Limited Dependent Variable and Panel Data Models, Cambridge University Press.
Anthony Davies and Kajal Lahiri
*Earlier versions of this paper were presented at the 5th. Biennial International Conference on Panel Data (Paris 1994), the 7th. World Congress of the Econometric Society (Tokyo 1995), and at the 17th. International Forecasting Symposium (Istanbul 1996). We thank G. S. Maddala, Carl Bonham, Michael Keane, Steven McNees, Hashem Pesaran, Christopher Sims, and Victor Zarnowitz for many comments and suggestions. We alone are responsible for any errors and shortcomings.
In this article, we develop a general econometric framework for analyzing the ASA-NBER expectations survey of professional forecasters. The framework allows for a complex correlation among forecast errors made by different individuals, for different targets, and at different horizons to be expressed as a function of a few fundamental parameters, and provides a more comprehensive tests for the Rational Expectations hypothesis. More importantly, this model formulation enables us to obtain direct measures of aggregate shocks and their volatility prior to the series being realized. The inflation forecasts derived from the ASA-NBER surveys are not found to be rational.
Return to Cline & Davies Research Alliance