Report

Deconstructing the yield curve

Abstract: We introduce a novel nonparametric bootstrap for assets with a finite maturity structure such as the nominal yield curve. We analyze the properties of our resampling procedure for inference on bond return predictability. Our method is asymptotically valid and robust to general forms of time and cross-sectional dependence; moreover, it exhibits excellent finite-sample properties. We demonstrate the applicability of our results in two empirical exercises: first, we show that a proxy for equity market tail risk predicts bond returns beyond yield curve factors; second, we provide a bootstrap bias correction and confidence intervals for yield-based probability of recession models.

Keywords: term structure of interest rates; factor models; principal components; bond risk premiums; resampling-based inference; predictive regression of bond returns;

JEL Classification: C15; C58; G10; G12;

Bibliographic Information

Provider: Federal Reserve Bank of New York

Part of Series: Staff Reports

Publication Date: 2019-04-01

Number: 884

Note: Revised January 2022.