Extreme Yield Movements and Dynamic Density Forecasts for the Term Structure of Interest Rates

In this paper, we utilize the dynamic Nelson-Siegel model to forecast the joint density of changes in the term structure of interest rates. We specify a flexible model for the factor dynamics and their dependence structure using time-varying copulas, thus allowing for departure from the normality assumption typically adopted in the literature. Using US Treasury yields for the period January 1970 to December 2009, our in-sample results show statistically significant gains due to allowing for a time-varying dependence structure which permits extreme factor comovements. In addition, allowing for time variation in the dependence structure significantly enhances the model's ability to forecast the joint density of the term structure. The latter results are based on Diebold and Mariano (1995) pairwise predictive ability tests, and the model confidence set of Hansen et al. (2011)