Structural Break Tests Robust to Regression Misspecification

Alaa Abi Morshed, E. Andreou, Otilia Boldea

Research output: Working paperDiscussion paperOther research output

Abstract

Structural break tests developed in the literature for regression models are sensitive to model misspecification. We show - analytically and through simulations - that the sup Wald test for breaks in the conditional mean and variance of a time series process exhibits severe size distortions when the conditional mean dynamics are misspecified. We also show that the sup Wald test for breaks in the unconditional mean and variance does not have the same size distortions, yet benefits from similar power to its conditional counterpart.
Hence, we propose using it as an alternative and complementary test for breaks. While the conditional tests based on dynamic regression models detect breaks in the mean and variance of the US unemployment growth and interest rate growth series around the Great Moderation, the evidence for these breaks disappears when using the unconditional tests. Therefore, there is no evidence of long-run mean or volatility shifts in unemployment growth and interest rate growth.
LanguageEnglish
Place of PublicationTilbug
PublisherCentER, Center for Economic Research
Number of pages39
Volume2016-019
Publication statusPublished - 2 May 2016

Publication series

NameCentER Discussion Paper
Volume2016-019

Fingerprint

Misspecification
Structural break tests
Unemployment
Wald test
Interest rates
Size distortion
Regression model
Great moderation
Simulation
Model misspecification

Keywords

  • structural change
  • sup Wald test
  • dynamic misspecification

Cite this

Abi Morshed, A., Andreou, E., & Boldea, O. (2016). Structural Break Tests Robust to Regression Misspecification. (CentER Discussion Paper; Vol. 2016-019). Tilbug: CentER, Center for Economic Research.
Abi Morshed, Alaa ; Andreou, E. ; Boldea, Otilia. / Structural Break Tests Robust to Regression Misspecification. Tilbug : CentER, Center for Economic Research, 2016. (CentER Discussion Paper).
@techreport{3b21f21c2cef49d7bb9bad81f5ee71f2,
title = "Structural Break Tests Robust to Regression Misspecification",
abstract = "Structural break tests developed in the literature for regression models are sensitive to model misspecification. We show - analytically and through simulations - that the sup Wald test for breaks in the conditional mean and variance of a time series process exhibits severe size distortions when the conditional mean dynamics are misspecified. We also show that the sup Wald test for breaks in the unconditional mean and variance does not have the same size distortions, yet benefits from similar power to its conditional counterpart.Hence, we propose using it as an alternative and complementary test for breaks. While the conditional tests based on dynamic regression models detect breaks in the mean and variance of the US unemployment growth and interest rate growth series around the Great Moderation, the evidence for these breaks disappears when using the unconditional tests. Therefore, there is no evidence of long-run mean or volatility shifts in unemployment growth and interest rate growth.",
keywords = "structural change, sup Wald test, dynamic misspecification",
author = "{Abi Morshed}, Alaa and E. Andreou and Otilia Boldea",
year = "2016",
month = "5",
day = "2",
language = "English",
volume = "2016-019",
series = "CentER Discussion Paper",
publisher = "CentER, Center for Economic Research",
type = "WorkingPaper",
institution = "CentER, Center for Economic Research",

}

Abi Morshed, A, Andreou, E & Boldea, O 2016 'Structural Break Tests Robust to Regression Misspecification' CentER Discussion Paper, vol. 2016-019, CentER, Center for Economic Research, Tilbug.

Structural Break Tests Robust to Regression Misspecification. / Abi Morshed, Alaa; Andreou, E.; Boldea, Otilia.

Tilbug : CentER, Center for Economic Research, 2016. (CentER Discussion Paper; Vol. 2016-019).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Structural Break Tests Robust to Regression Misspecification

AU - Abi Morshed, Alaa

AU - Andreou, E.

AU - Boldea, Otilia

PY - 2016/5/2

Y1 - 2016/5/2

N2 - Structural break tests developed in the literature for regression models are sensitive to model misspecification. We show - analytically and through simulations - that the sup Wald test for breaks in the conditional mean and variance of a time series process exhibits severe size distortions when the conditional mean dynamics are misspecified. We also show that the sup Wald test for breaks in the unconditional mean and variance does not have the same size distortions, yet benefits from similar power to its conditional counterpart.Hence, we propose using it as an alternative and complementary test for breaks. While the conditional tests based on dynamic regression models detect breaks in the mean and variance of the US unemployment growth and interest rate growth series around the Great Moderation, the evidence for these breaks disappears when using the unconditional tests. Therefore, there is no evidence of long-run mean or volatility shifts in unemployment growth and interest rate growth.

AB - Structural break tests developed in the literature for regression models are sensitive to model misspecification. We show - analytically and through simulations - that the sup Wald test for breaks in the conditional mean and variance of a time series process exhibits severe size distortions when the conditional mean dynamics are misspecified. We also show that the sup Wald test for breaks in the unconditional mean and variance does not have the same size distortions, yet benefits from similar power to its conditional counterpart.Hence, we propose using it as an alternative and complementary test for breaks. While the conditional tests based on dynamic regression models detect breaks in the mean and variance of the US unemployment growth and interest rate growth series around the Great Moderation, the evidence for these breaks disappears when using the unconditional tests. Therefore, there is no evidence of long-run mean or volatility shifts in unemployment growth and interest rate growth.

KW - structural change

KW - sup Wald test

KW - dynamic misspecification

M3 - Discussion paper

VL - 2016-019

T3 - CentER Discussion Paper

BT - Structural Break Tests Robust to Regression Misspecification

PB - CentER, Center for Economic Research

CY - Tilbug

ER -

Abi Morshed A, Andreou E, Boldea O. Structural Break Tests Robust to Regression Misspecification. Tilbug: CentER, Center for Economic Research. 2016 May 2. (CentER Discussion Paper).