Home > Research > Publications & Outputs > Testing Stationarity with Unobserved Components...
View graph of relations

Testing Stationarity with Unobserved Components Models

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Testing Stationarity with Unobserved Components Models. / Morley, James; Panovska, Irina B.; Sinclair, Tara M.
In: Macroeconomic Dynamics, Vol. 21, No. 1, 01.2017, p. 160-182.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Morley, J, Panovska, IB & Sinclair, TM 2017, 'Testing Stationarity with Unobserved Components Models', Macroeconomic Dynamics, vol. 21, no. 1, pp. 160-182. https://doi.org/10.1017/S1365100515000437

APA

Morley, J., Panovska, I. B., & Sinclair, T. M. (2017). Testing Stationarity with Unobserved Components Models. Macroeconomic Dynamics, 21(1), 160-182. https://doi.org/10.1017/S1365100515000437

Vancouver

Morley J, Panovska IB, Sinclair TM. Testing Stationarity with Unobserved Components Models. Macroeconomic Dynamics. 2017 Jan;21(1):160-182. Epub 2016 Mar 8. doi: 10.1017/S1365100515000437

Author

Morley, James ; Panovska, Irina B. ; Sinclair, Tara M. / Testing Stationarity with Unobserved Components Models. In: Macroeconomic Dynamics. 2017 ; Vol. 21, No. 1. pp. 160-182.

Bibtex

@article{4480be677117489d83e60b17be0c84d0,
title = "Testing Stationarity with Unobserved Components Models",
abstract = "In the aftermath of the global financial crisis, competing measures of the trend in macroeconomic variables such as U.S. real GDP have featured prominently in policy debates. A key question is whether large shocks to macroeconomic variables will have permanent effects—i.e., in econometric terms, do the data contain stochastic trends? Unobserved-components models provide a convenient way to estimate stochastic trends for time series data, with their existence typically motivated by stationarity tests that allow at most a deterministic trend under the null hypothesis. However, given the small sample sizes available for most macroeconomic variables, standard Lagrange multiplier tests of stationarity will perform poorly when the data are highly persistent. To address this problem, we propose the use of a likelihood ratio test of stationarity based directly on the unobserved-components models used in estimation of stochastic trends. We demonstrate that a bootstrap version of this test has far better small-sample properties for empirically relevant data-generating processes than bootstrap versions of the standard Lagrange multiplier tests. An application to U.S. real GDP produces stronger support for the presence of large permanent shocks using the likelihood ratio test than using the standard tests.",
keywords = "Stationarity Test, Likelihood ratio, Unobserved Components, Parametric Bootstrap, Monte Carlo Simulation, Small-Sample Inference",
author = "James Morley and Panovska, {Irina B.} and Sinclair, {Tara M.}",
year = "2017",
month = jan,
doi = "10.1017/S1365100515000437",
language = "English",
volume = "21",
pages = "160--182",
journal = "Macroeconomic Dynamics",
issn = "1365-1005",
publisher = "Cambridge University Press",
number = "1",

}

RIS

TY - JOUR

T1 - Testing Stationarity with Unobserved Components Models

AU - Morley, James

AU - Panovska, Irina B.

AU - Sinclair, Tara M.

PY - 2017/1

Y1 - 2017/1

N2 - In the aftermath of the global financial crisis, competing measures of the trend in macroeconomic variables such as U.S. real GDP have featured prominently in policy debates. A key question is whether large shocks to macroeconomic variables will have permanent effects—i.e., in econometric terms, do the data contain stochastic trends? Unobserved-components models provide a convenient way to estimate stochastic trends for time series data, with their existence typically motivated by stationarity tests that allow at most a deterministic trend under the null hypothesis. However, given the small sample sizes available for most macroeconomic variables, standard Lagrange multiplier tests of stationarity will perform poorly when the data are highly persistent. To address this problem, we propose the use of a likelihood ratio test of stationarity based directly on the unobserved-components models used in estimation of stochastic trends. We demonstrate that a bootstrap version of this test has far better small-sample properties for empirically relevant data-generating processes than bootstrap versions of the standard Lagrange multiplier tests. An application to U.S. real GDP produces stronger support for the presence of large permanent shocks using the likelihood ratio test than using the standard tests.

AB - In the aftermath of the global financial crisis, competing measures of the trend in macroeconomic variables such as U.S. real GDP have featured prominently in policy debates. A key question is whether large shocks to macroeconomic variables will have permanent effects—i.e., in econometric terms, do the data contain stochastic trends? Unobserved-components models provide a convenient way to estimate stochastic trends for time series data, with their existence typically motivated by stationarity tests that allow at most a deterministic trend under the null hypothesis. However, given the small sample sizes available for most macroeconomic variables, standard Lagrange multiplier tests of stationarity will perform poorly when the data are highly persistent. To address this problem, we propose the use of a likelihood ratio test of stationarity based directly on the unobserved-components models used in estimation of stochastic trends. We demonstrate that a bootstrap version of this test has far better small-sample properties for empirically relevant data-generating processes than bootstrap versions of the standard Lagrange multiplier tests. An application to U.S. real GDP produces stronger support for the presence of large permanent shocks using the likelihood ratio test than using the standard tests.

KW - Stationarity Test

KW - Likelihood ratio

KW - Unobserved Components

KW - Parametric Bootstrap

KW - Monte Carlo Simulation

KW - Small-Sample Inference

U2 - 10.1017/S1365100515000437

DO - 10.1017/S1365100515000437

M3 - Journal article

VL - 21

SP - 160

EP - 182

JO - Macroeconomic Dynamics

JF - Macroeconomic Dynamics

SN - 1365-1005

IS - 1

ER -