Premium
Linear and Non‐Linear Granger Causality Between Short‐Term and Long‐Term Interest Rates: A Rolling Window Strategy
Author(s) -
Rahimi Azadeh,
Chu Ba M.,
Lavoie Marc
Publication year - 2017
Publication title -
metroeconomica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.256
H-Index - 29
eISSN - 1467-999X
pISSN - 0026-1386
DOI - 10.1111/meca.12148
Subject(s) - granger causality , causality (physics) , term (time) , econometrics , economics , linear model , government bond , interest rate , mathematics , statistics , macroeconomics , physics , quantum mechanics
In this article, a rolling window strategy is used to detect the linear and non‐linear Granger causality relationships between the U.S. federal funds rate and the 10‐year government bond rate, during different time horizons, investigating whether these causalities change with the passing of time. For linear Granger causality tests, we apply the Toda and Yamamoto ([Toda, H. Y., 1995]) approach and for non‐linear ones we use a non‐linear Granger causality test introduced by Diks and Panchenko ([Diks, C., 2006]). Our findings show that during nearly all time periods there is a significant two‐way Granger causality relationship between these two interest rates.