A1533
Title: On unspanned latent risks in dynamic term structure models
Authors: Tomasz Dubiel-Teleszynski - University of Liechtenstein (Liechtenstein) [presenting]
Konstantinos Kalogeropoulos - London School of Economics (United Kingdom)
Nikolaos Karouzakis - University of Sussex (United Kingdom)
Abstract: A parsimonious class of arbitrage-free yields-only Dynamic Term Structure Models (DTSMs) with unspanned latent risks is proposed. We develop an efficient Sequential Monte Carlo (SMC) inferential and prediction scheme that guarantees joint identification of parameters and latent states and takes into account all relevant uncertainties. We use the developed setup to explore the out-of-sample statistical and economic evidence of bond return predictability from the perspective of a real-time Bayesian investor seeking to forecast excess bond returns and maximise her utility. We find that latent factors contain significant predictive power above and beyond the yield curve, offering significant improvement to the out-of-sample predictive performance of models. Most importantly, they exploit information hidden from the yield curve and generate significant utility gains, out-of-sample. Macro-financial linkages are also explored. The hidden component associated with slope risk is countercyclical and links with real activity.
