FORECASTING THE YIELD CURVE USING PRIORS FROM NO-ARBITRAGE AFFINE TERM STRUCTURE MODELS

Authors

  • Andrea Carriero

    1. University of London, U.K.
    Search for more papers by this author
    • This article has greatly benefited from several comments and detailed suggestions by three anonymous referees and by Frank Schorfheide. I also thank seminar participants at the University of Bristol, at the EC2 meeting in Faro, and at the London-OxBridge Time Series Workshop in LSE. Errors are mine. Please address correspondence to: Andrea Carriero, Department of Economics, Queen Mary, University of London, Mile End Road, London E1 4NS, U.K. Phone: +44-20-7882-8050. E-mail: a.carriero@qmul.ac.uk.


  • Manuscript received September 2007; revised October 2008.

Abstract

I propose a strategy for forecasting the term structure of interest rates that may produce significant gains in predictive accuracy. The key idea is to use the restrictions implied by Gaussian, no-arbitrage, affine term structure models on a vector autoregression as prior information instead of imposing the restrictions dogmatically. This allows us to account for possible model misspecification. We use the proposed method to forecast a system of five U.S. yields up to 12 months ahead, and we find it provides significant gains in forecast accuracy.

Ancillary