Title: Forecasting the term structure of interest rates with potentially misspecified models
Authors: Yunjong Eo - University of Sydney (Australia) [presenting]
Kyu Ho Kang - Korea University (Korea, South)
Abstract: The predictive gains of a Markov-switching mixture of three individual bond yield predictions is assessed: namely, the dynamic Nelson-Siegel model (DNS), the arbitrage-free Nelson-Siegel model, and the random walk (RW) model as a benchmark. Despite the popularity of these three frameworks, none of them dominates the others across all maturities and forecast horizons. This fact indicates that the models are potentially misspecified. We investigate whether combining the possibly misspecified models in a linear form helps improve predictive accuracy. To do this, we evaluate the out-of-sample forecasts of the mixture models compared to the individual models. Our findings provide strong evidence that model combination can be a better option than selecting one of the alternative models.