Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter (peer-reviewed)
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter (peer-reviewed)
}
TY - CHAP
T1 - Testing for a Markov-Switching Mean in Serially Correlated Data
AU - Morley, James
AU - Rabah, Zohra
PY - 2014
Y1 - 2014
N2 - When testing for Markov switching in mean or intercept of an autoregressiveprocess, it is important to allow for serial correlation under the nullhypothesis of linearity. Otherwise, a rejection of linearity could merely reflectmisspecification of the persistence properties of the data, rather than any inherent nonlinearity. However, Monte Carlo analysis reveals that the Carrasco, Hu, and Ploberger (Optimal test for Markov Switching parameters, conditionally accepted at Econometrica, 2012) test for Markov switching has low power for empirically relevant data-generating processes when allowing for serial correlation under the null. By contrast, a parametric bootstrap likelihood ratio test of Markov switching has higher power in the same setting. Correspondingly, the bootstrap likelihood ratio test provides stronger support for a Markov-switching mean in an application to an autoregressive model of quarterly US real GDP growth.
AB - When testing for Markov switching in mean or intercept of an autoregressiveprocess, it is important to allow for serial correlation under the nullhypothesis of linearity. Otherwise, a rejection of linearity could merely reflectmisspecification of the persistence properties of the data, rather than any inherent nonlinearity. However, Monte Carlo analysis reveals that the Carrasco, Hu, and Ploberger (Optimal test for Markov Switching parameters, conditionally accepted at Econometrica, 2012) test for Markov switching has low power for empirically relevant data-generating processes when allowing for serial correlation under the null. By contrast, a parametric bootstrap likelihood ratio test of Markov switching has higher power in the same setting. Correspondingly, the bootstrap likelihood ratio test provides stronger support for a Markov-switching mean in an application to an autoregressive model of quarterly US real GDP growth.
KW - Nonlinearity tests
KW - Autoregressive processes
KW - Markov switching
KW - Parametric bootstrap
KW - Real GDP dynamics
U2 - 10.1007/978-1-4614-8060-0_5
DO - 10.1007/978-1-4614-8060-0_5
M3 - Chapter (peer-reviewed)
SN - 9781461480594
SP - 85
EP - 97
BT - Recent Advances in Estimating Nonlinear Models
A2 - Ma, Jun
A2 - Wohar, Mark
PB - Springer
CY - New York
ER -