Are Bond Returns Predictable with Real-Time Macro Data?
52 Pages Posted: 23 Jan 2018 Last revised: 5 Mar 2019
Date Written: March 2019
We reaffirm the stylized fact that bond risk premia are time-varying with macroeconomic condition, even with real-time macro data instead of commonly used final revised data. While real-time data are noisier and render standard forecasts insignificant, we find that, with four efficient target-driven methods, they still contain enough information to predict bond returns significantly both in- and out-of-sample. The predictability can also yield substantial economic value to a mean-variance investor. Moreover, the factors extracted from real-time data predict future macroeconomic condition. Consistent with asset pricing theory, the predicted bond returns are countercyclical.
Keywords: Bond Return Predictability, Real Time Macro Data, Vintage, PCA, Big Data, Machine Learning
JEL Classification: C22, C53, G11, G12, G17
Suggested Citation: Suggested Citation