Distribution theory for unit root tests with conditional heteroskedasticity

Research output: Contribution to journalArticle

74 Citations (Scopus)

Abstract

This paper explores the asymptotic distribution theory of autoregressive (AR) unit root tests where the error follows a generalized autoregressive conditional heteroskedastic (GARCH) process. The proposed unit root test is based on maximum likelihood estimation, which estimates the AR unit root and the GARCH parameters jointly. The asymptotic distribution of the t-statistic for the AR unit root is a mixture of the Dickey-Fuller t-distribution and the standard normal, with the relative weight depending on the magnitude of the GARCH effect and the fourth moment of the standardized errors. As the GARCH effect increases, the power of the tests improves significantly. These results show that significant power gains emerge from the joint estimation rather than relying on the conventional ADF test which ignores the heteroskedasticity in the data.

Original languageEnglish
Pages (from-to)113-144
Number of pages32
JournalJournal of Econometrics
Volume91
Issue number1
Publication statusPublished - 1999 Jul 1
Externally publishedYes

Fingerprint

Conditional Heteroskedasticity
Unit Root Tests
Distribution Theory
Maximum likelihood estimation
Statistics
Unit Root
Asymptotic distribution
Heteroskedasticity
t-distribution
Asymptotic Theory
Unit root tests
Conditional heteroskedasticity
Maximum Likelihood Estimation
Statistic
Moment

Keywords

  • GARCH
  • MLE
  • Power
  • Unit root

ASJC Scopus subject areas

  • Economics and Econometrics
  • Finance
  • Statistics and Probability

Cite this

Distribution theory for unit root tests with conditional heteroskedasticity. / Seo, Byeongseon.

In: Journal of Econometrics, Vol. 91, No. 1, 01.07.1999, p. 113-144.

Research output: Contribution to journalArticle

@article{d7a510c7eda44fa3b87cfd360b5946bc,
title = "Distribution theory for unit root tests with conditional heteroskedasticity",
abstract = "This paper explores the asymptotic distribution theory of autoregressive (AR) unit root tests where the error follows a generalized autoregressive conditional heteroskedastic (GARCH) process. The proposed unit root test is based on maximum likelihood estimation, which estimates the AR unit root and the GARCH parameters jointly. The asymptotic distribution of the t-statistic for the AR unit root is a mixture of the Dickey-Fuller t-distribution and the standard normal, with the relative weight depending on the magnitude of the GARCH effect and the fourth moment of the standardized errors. As the GARCH effect increases, the power of the tests improves significantly. These results show that significant power gains emerge from the joint estimation rather than relying on the conventional ADF test which ignores the heteroskedasticity in the data.",
keywords = "GARCH, MLE, Power, Unit root",
author = "Byeongseon Seo",
year = "1999",
month = "7",
day = "1",
language = "English",
volume = "91",
pages = "113--144",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - Distribution theory for unit root tests with conditional heteroskedasticity

AU - Seo, Byeongseon

PY - 1999/7/1

Y1 - 1999/7/1

N2 - This paper explores the asymptotic distribution theory of autoregressive (AR) unit root tests where the error follows a generalized autoregressive conditional heteroskedastic (GARCH) process. The proposed unit root test is based on maximum likelihood estimation, which estimates the AR unit root and the GARCH parameters jointly. The asymptotic distribution of the t-statistic for the AR unit root is a mixture of the Dickey-Fuller t-distribution and the standard normal, with the relative weight depending on the magnitude of the GARCH effect and the fourth moment of the standardized errors. As the GARCH effect increases, the power of the tests improves significantly. These results show that significant power gains emerge from the joint estimation rather than relying on the conventional ADF test which ignores the heteroskedasticity in the data.

AB - This paper explores the asymptotic distribution theory of autoregressive (AR) unit root tests where the error follows a generalized autoregressive conditional heteroskedastic (GARCH) process. The proposed unit root test is based on maximum likelihood estimation, which estimates the AR unit root and the GARCH parameters jointly. The asymptotic distribution of the t-statistic for the AR unit root is a mixture of the Dickey-Fuller t-distribution and the standard normal, with the relative weight depending on the magnitude of the GARCH effect and the fourth moment of the standardized errors. As the GARCH effect increases, the power of the tests improves significantly. These results show that significant power gains emerge from the joint estimation rather than relying on the conventional ADF test which ignores the heteroskedasticity in the data.

KW - GARCH

KW - MLE

KW - Power

KW - Unit root

UR - http://www.scopus.com/inward/record.url?scp=0005837363&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0005837363&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0005837363

VL - 91

SP - 113

EP - 144

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1

ER -