TY - JOUR
T1 - The effects of conventional and unconventional monetary policy on forecasting the yield curve
AU - Eo, Yunjong
AU - Kang, Kyu Ho
N1 - Funding Information:
We thank two anonymous referees, Gianni Amisano, James Bullard, Sung Je Byun, Cem Cakmakh, Marcelle Chauvet, Alex Chudik, Graham Elliott, Domenico Giannone, Massimo Guidolin, Leo Krippner, Tae-hwy Lee, Michael McCracken, James Morley, Barbara Rossi, Rodrigo Sekkel, Tara Sinclair, Dan Waggoner, Toshiaki Watanabe, Natassa Zervou, and seminar and conference participants at Texas A&M University, the Federal Reserve Bank of Dallas, the Reserve Bank of New Zealand, University of California Riverside, Hitotsubashi University, the Bank of Korea, the Narodowy Bank Polski Workshop on Forecasting, the International Association for Applied Econometrics conference, the Society for Nonlinear Dynamics and Econometrics Conference, Recent Developments in Financial Econometrics and Applications, EFaB@Bayes 250 Workshop, International Conference on Computational and Financial Econometrics, and Econometric Society Australasian Meeting for useful feedback. Tom Cusbert and Andrew Gaffney provided invaluable research assistance. This research is supported by the FASS Strategic Research Scheme from University of Sydney. All remaining errors are our own. This article is a substantially revised version of an earlier work previously circulated under the title ?Forecasting the term structure of interest rates with potentially misspecified models.?
Publisher Copyright:
© 2019
PY - 2020/2
Y1 - 2020/2
N2 - We investigate how conventional and unconventional monetary policies affect the dynamics of the yield curve by assessing the performance of individual yield curve models and their mixtures. Out-of-sample forecasts for U.S. bond yields show that the arbitrage-free Nelson–Siegel model and its mixtures with other models perform well in the period of conventional monetary policy, whereas the random walk model outperforms all the other models in the period of unconventional monetary policy. The diminished role of the no-arbitrage restriction in forecasting the yield curve since 2009 can be attributed to unconventional monetary policy, which resulted in low correlations between short- and long-term bond yields and little variation in the short-term rates. During the period of the maturity extension program in 2011–2012, the superiority of the random walk forecasts is more pronounced, reinforcing our finding that the monetary policy framework affects yield curve forecast accuracy.
AB - We investigate how conventional and unconventional monetary policies affect the dynamics of the yield curve by assessing the performance of individual yield curve models and their mixtures. Out-of-sample forecasts for U.S. bond yields show that the arbitrage-free Nelson–Siegel model and its mixtures with other models perform well in the period of conventional monetary policy, whereas the random walk model outperforms all the other models in the period of unconventional monetary policy. The diminished role of the no-arbitrage restriction in forecasting the yield curve since 2009 can be attributed to unconventional monetary policy, which resulted in low correlations between short- and long-term bond yields and little variation in the short-term rates. During the period of the maturity extension program in 2011–2012, the superiority of the random walk forecasts is more pronounced, reinforcing our finding that the monetary policy framework affects yield curve forecast accuracy.
KW - Arbitrage-free term structure model
KW - Dynamic Nelson–Siegel model
KW - Markov-switching mixture
KW - Operation twist
KW - Random walk model
UR - http://www.scopus.com/inward/record.url?scp=85076146634&partnerID=8YFLogxK
U2 - 10.1016/j.jedc.2019.103812
DO - 10.1016/j.jedc.2019.103812
M3 - Article
AN - SCOPUS:85076146634
SN - 0165-1889
VL - 111
JO - Journal of Economic Dynamics and Control
JF - Journal of Economic Dynamics and Control
M1 - 103812
ER -