Forecasting Bond Yields with Segmented Term Structure Models
51 Pages Posted: 30 Aug 2013 Last revised: 14 Dec 2016
Date Written: December 9, 2016
Inspired by the preferred-habitat theory, we propose parametric interest rate models that split the term structure into segments. The proposed models are compared to successful term structure benchmarks based on out-of-sample forecasting exercises using US Treasury data. We show that segmentation can improve long-horizon term structure forecasts when compared to non-segmentation. Additionally, introducing cointegration in latent factor dynamics of segmented models makes them particularly strong to forecast short-maturity yields. Better forecasting is justified by the segmented models' ability to accommodate idiosyncratic shocks in the cross-section of yields.
Keywords: Preferred-Habitat Theory, Error Correction Models, Model Selection, Exponential Splines, Local Shocks
JEL Classification: G51, G52, G53
Suggested Citation: Suggested Citation