Yield spread selection in predicting recession probabilities

Jaehyuk Choi, Desheng Ge, Kyu Ho Kang, Sungbin Sohn

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

The literature on using yield curves to forecast recessions customarily uses 10-year–3-month Treasury yield spread without verification on the pair selection. This study investigates whether the predictive ability of spread can be improved by letting a machine learning algorithm identify the best maturity pair and coefficients. Our comprehensive analysis shows that, despite the likelihood gain, the machine learning approach does not significantly improve prediction, owing to the estimation error. This is robust to the forecasting horizon, control variable, sample period, and oversampling of the recession observations. Our finding supports the use of the 10-year–3-month spread.

Original languageEnglish
Pages (from-to)1772-1785
Number of pages14
JournalJournal of Forecasting
Volume42
Issue number7
DOIs
Publication statusPublished - 2023 Nov

Bibliographical note

Funding Information:
This work was supported by the Korea University (K2302511) and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF‐2022M3J6A1063595). Sungbin Sohn acknowledges financial support from the Sogang University Research Grant of 2022 (No. 202210030.01).

Publisher Copyright:
© 2023 John Wiley & Sons Ltd.

Keywords

  • density forecasting
  • estimation risk
  • machine learning
  • yield curve

ASJC Scopus subject areas

  • Economics and Econometrics
  • Computer Science Applications
  • Statistics, Probability and Uncertainty
  • Modelling and Simulation
  • Strategy and Management
  • Management Science and Operations Research

Fingerprint

Dive into the research topics of 'Yield spread selection in predicting recession probabilities'. Together they form a unique fingerprint.

Cite this