In this paper we propose a strategy for forecasting the term structure of interest rates which may produce significant gains in predictive accuracy. The key idea is to use the restrictions implied by Affine Term Structure Models (ATSM) on a vector autoregression (VAR) as prior information rather than imposing them dogmatically. This allows to account for possible model misspecification. We apply the method to a system of five US yields, and we find that the gains in predictive accuracy can be substantial. In particular, for horizons longer than 1-step ahead, our proposed method produces systematically better forecasts than those obtained by using a pure ATSM or an unrestricted VAR, and it also outperforms very competitive benchmarks as the Minnesota prior, the Diebold-Li (2006) model, and the random walk.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
file. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Publisher Info
Paper provided by Queen Mary, University of London, Department of Economics in its series Working Papers with number
612.
Find related papers by JEL classification: C11 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods: General - - - Bayesian Analysis C53 - Mathematical and Quantitative Methods - - Econometric Modeling - - - Forecasting and Other Model Applications E43 - Macroeconomics and Monetary Economics - - Money and Interest Rates - - - Determination of Interest Rates; Term Structure of Interest Rates E47 - Macroeconomics and Monetary Economics - - Money and Interest Rates - - - Forecasting and Simulation
This paper has been announced in the following NEP Reports:
Did you know? You can create a compilation of all publications of a group of people, say alumni of a program, your students or memers of an association.