Get access

ESTIMATING DYNAMIC EULER EQUATIONS WITH MULTIVARIATE PROFESSIONAL FORECASTS

Authors

  • GREGOR W. SMITH,

    1. Smith: Professor, Department of Economics, Queen's University, Kingston, Ontario K7L 3N6, Canada. Phone 1-613-533-6659, Fax 1-613-533-6668, E-mail smithgw@econ.queensu.ca
    Search for more papers by this author
  • JAMES YETMAN

    1. Yetman: Senior Economist, Bank for International Settlements, Hong Kong, China. Phone 852-9271-7354, Fax 852-2878-7123, E-mail james.yetman@bis.org
    Search for more papers by this author
    • Earlier versions were entitled “The Curse of Irving Fisher (Professional Forecasters' Version).” We thank the Social Sciences and Humanities Research Council of Canada and the Bank of Canada research fellowship program for support of this research. The opinions are the authors' alone and are not those of the Bank of Canada or the Bank for International Settlements. We thank Frank Diebold, Dean Croushore, John Galbraith, Kim Huynh, Atsushi Inoue, Vadim Marmer, Angelo Melino, James Nason, Jim Vere, Casey Warman, Jonathan Wright, and seminar participants at Indiana University, Simon Fraser University, Monash University, the University of Victoria, Ryerson University, McGill University, and the Canadian Macroeconomics Study Group for helpful comments. Two referees and a co-editor of this journal provided very helpful criticism that improved the paper.


Abstract

Dynamic Euler equations restrict multivariate forecasts and so can be estimated and tested using the predictions of professional forecasters. We illustrate this novel, empirical method by studying the links between forecasts of U.S. nominal interest rates, inflation, and real consumption growth since 1981. Using forecast data for both returns and macroeconomic fundamentals exploits the complete panel of forecasts from the Survey of Professional Forecasters, which yields 3,400 observations, many more than the 117 quarterly time-series observations. Harnessing the full panel enhances precision in testing asset-pricing models and may avoid aggregation bias. We find clear evidence for the Fisher effect but mixed evidence of a relationship between expectations of real interest rates and real consumption growth. (JEL E17, E21, E43)

Ancillary