A robust test for autocorrelation in the presence of a structural break in variance

Hyeong Ho Mun, Eun Young Shim, Tae Hwan Kim

Research output: Contribution to journalArticle

Abstract

It has been known that when there is a break in the variance (unconditional heteroskedasticity) of the error term in linear regression models, a routine application of the Lagrange multiplier (LM) test for autocorrelation can cause potentially significant size distortions. We propose a new test for autocorrelation that is robust in the presence of a break in variance. The proposed test is a modified LM test based on a generalized least squares regression. Monte Carlo simulations show that the new test performs well in finite samples and it is especially comparable to other existing heteroskedasticity-robust tests in terms of size, and much better in terms of power.

Original languageEnglish
Pages (from-to)1552-1562
Number of pages11
JournalJournal of Statistical Computation and Simulation
Volume84
Issue number7
DOIs
Publication statusPublished - 2014 Jul

Fingerprint

Structural Breaks
Robust Tests
Lagrange multipliers
Autocorrelation
Lagrange multiplier Test
Heteroskedasticity
Linear regression
Size Distortion
Generalized Least Squares
Least Squares Regression
Error term
Linear Regression Model
Monte Carlo Simulation
Structural breaks
Monte Carlo simulation
Lagrange multiplier test

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modelling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

@article{0f0cce6d2e41480f83f467df1a1b3786,
title = "A robust test for autocorrelation in the presence of a structural break in variance",
abstract = "It has been known that when there is a break in the variance (unconditional heteroskedasticity) of the error term in linear regression models, a routine application of the Lagrange multiplier (LM) test for autocorrelation can cause potentially significant size distortions. We propose a new test for autocorrelation that is robust in the presence of a break in variance. The proposed test is a modified LM test based on a generalized least squares regression. Monte Carlo simulations show that the new test performs well in finite samples and it is especially comparable to other existing heteroskedasticity-robust tests in terms of size, and much better in terms of power.",
author = "Mun, {Hyeong Ho} and Shim, {Eun Young} and Kim, {Tae Hwan}",
year = "2014",
month = "7",
doi = "10.1080/00949655.2012.754027",
language = "English",
volume = "84",
pages = "1552--1562",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "7",

}

A robust test for autocorrelation in the presence of a structural break in variance. / Mun, Hyeong Ho; Shim, Eun Young; Kim, Tae Hwan.

In: Journal of Statistical Computation and Simulation, Vol. 84, No. 7, 07.2014, p. 1552-1562.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A robust test for autocorrelation in the presence of a structural break in variance

AU - Mun, Hyeong Ho

AU - Shim, Eun Young

AU - Kim, Tae Hwan

PY - 2014/7

Y1 - 2014/7

N2 - It has been known that when there is a break in the variance (unconditional heteroskedasticity) of the error term in linear regression models, a routine application of the Lagrange multiplier (LM) test for autocorrelation can cause potentially significant size distortions. We propose a new test for autocorrelation that is robust in the presence of a break in variance. The proposed test is a modified LM test based on a generalized least squares regression. Monte Carlo simulations show that the new test performs well in finite samples and it is especially comparable to other existing heteroskedasticity-robust tests in terms of size, and much better in terms of power.

AB - It has been known that when there is a break in the variance (unconditional heteroskedasticity) of the error term in linear regression models, a routine application of the Lagrange multiplier (LM) test for autocorrelation can cause potentially significant size distortions. We propose a new test for autocorrelation that is robust in the presence of a break in variance. The proposed test is a modified LM test based on a generalized least squares regression. Monte Carlo simulations show that the new test performs well in finite samples and it is especially comparable to other existing heteroskedasticity-robust tests in terms of size, and much better in terms of power.

UR - http://www.scopus.com/inward/record.url?scp=84899427791&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899427791&partnerID=8YFLogxK

U2 - 10.1080/00949655.2012.754027

DO - 10.1080/00949655.2012.754027

M3 - Article

AN - SCOPUS:84899427791

VL - 84

SP - 1552

EP - 1562

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 7

ER -