Home > Research > Publications & Outputs > Testing for a Markov-Switching Mean in Serially...

Links

Text available via DOI:

View graph of relations

Testing for a Markov-Switching Mean in Serially Correlated Data

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNChapter (peer-reviewed)

Published

Standard

Testing for a Markov-Switching Mean in Serially Correlated Data. / Morley, James; Rabah, Zohra.
Recent Advances in Estimating Nonlinear Models : With Applications in Economics and Finance. ed. / Jun Ma; Mark Wohar. New York: Springer, 2014. p. 85-97.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNChapter (peer-reviewed)

Harvard

Morley, J & Rabah, Z 2014, Testing for a Markov-Switching Mean in Serially Correlated Data. in J Ma & M Wohar (eds), Recent Advances in Estimating Nonlinear Models : With Applications in Economics and Finance. Springer, New York, pp. 85-97. https://doi.org/10.1007/978-1-4614-8060-0_5

APA

Morley, J., & Rabah, Z. (2014). Testing for a Markov-Switching Mean in Serially Correlated Data. In J. Ma, & M. Wohar (Eds.), Recent Advances in Estimating Nonlinear Models : With Applications in Economics and Finance (pp. 85-97). Springer. https://doi.org/10.1007/978-1-4614-8060-0_5

Vancouver

Morley J, Rabah Z. Testing for a Markov-Switching Mean in Serially Correlated Data. In Ma J, Wohar M, editors, Recent Advances in Estimating Nonlinear Models : With Applications in Economics and Finance. New York: Springer. 2014. p. 85-97 Epub 2013 Aug 27. doi: 10.1007/978-1-4614-8060-0_5

Author

Morley, James ; Rabah, Zohra. / Testing for a Markov-Switching Mean in Serially Correlated Data. Recent Advances in Estimating Nonlinear Models : With Applications in Economics and Finance. editor / Jun Ma ; Mark Wohar. New York : Springer, 2014. pp. 85-97

Bibtex

@inbook{e6e3acc80c7d4dc9ac1b2e12717da6a4,
title = "Testing for a Markov-Switching Mean in Serially Correlated Data",
abstract = "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.",
keywords = "Nonlinearity tests, Autoregressive processes, Markov switching, Parametric bootstrap, Real GDP dynamics",
author = "James Morley and Zohra Rabah",
year = "2014",
doi = "10.1007/978-1-4614-8060-0_5",
language = "English",
isbn = "9781461480594",
pages = "85--97",
editor = "Jun Ma and Mark Wohar",
booktitle = "Recent Advances in Estimating Nonlinear Models",
publisher = "Springer",

}

RIS

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 -