Inference on structural breaks using information criteria

Alastair R. Hall, Denise R. Osborn, Nikolaos Sakkas

Research output: Contribution to journalArticle

10 Citations (Scopus)
97 Downloads (Pure)

Abstract

This paper investigates the usefulness of information criteria for inference on the number of structural breaks in a standard linear regression model. In particular, we propose a modified penalty function for such criteria, which implies each break is equivalent to estimation of three individual regression coefficients. A Monte Carlo analysis compares information criteria to sequential testing, with the modified Bayesian and Hannan-Quinn criteria performing well overall, for data-generating processes both without and with breaks. The methods are also used to examine changes in Euro area monetary policy between 1971 and 2007.
Original languageEnglish
Pages (from-to)54-81
JournalThe Manchester School
Volume81
Issue numberS3
Early online date29 Jul 2013
DOIs
Publication statusPublished - 2013

Fingerprint

Inference
Information criterion
Structural breaks
Penalty function
Data generating process
Monte Carlo analysis
Linear regression model
Testing
Usefulness
Coefficients
Monetary policy
Euro area

Cite this

Inference on structural breaks using information criteria. / Hall, Alastair R.; Osborn, Denise R.; Sakkas, Nikolaos.

In: The Manchester School, Vol. 81, No. S3, 2013, p. 54-81.

Research output: Contribution to journalArticle

Hall, Alastair R. ; Osborn, Denise R. ; Sakkas, Nikolaos. / Inference on structural breaks using information criteria. In: The Manchester School. 2013 ; Vol. 81, No. S3. pp. 54-81.
@article{618d7f1442954ee8ab8238274c361c10,
title = "Inference on structural breaks using information criteria",
abstract = "This paper investigates the usefulness of information criteria for inference on the number of structural breaks in a standard linear regression model. In particular, we propose a modified penalty function for such criteria, which implies each break is equivalent to estimation of three individual regression coefficients. A Monte Carlo analysis compares information criteria to sequential testing, with the modified Bayesian and Hannan-Quinn criteria performing well overall, for data-generating processes both without and with breaks. The methods are also used to examine changes in Euro area monetary policy between 1971 and 2007.",
author = "Hall, {Alastair R.} and Osborn, {Denise R.} and Nikolaos Sakkas",
year = "2013",
doi = "10.1111/manc.12017",
language = "English",
volume = "81",
pages = "54--81",
journal = "The Manchester School",
issn = "1467-9957",
publisher = "Wiley-Blackwell",
number = "S3",

}

TY - JOUR

T1 - Inference on structural breaks using information criteria

AU - Hall, Alastair R.

AU - Osborn, Denise R.

AU - Sakkas, Nikolaos

PY - 2013

Y1 - 2013

N2 - This paper investigates the usefulness of information criteria for inference on the number of structural breaks in a standard linear regression model. In particular, we propose a modified penalty function for such criteria, which implies each break is equivalent to estimation of three individual regression coefficients. A Monte Carlo analysis compares information criteria to sequential testing, with the modified Bayesian and Hannan-Quinn criteria performing well overall, for data-generating processes both without and with breaks. The methods are also used to examine changes in Euro area monetary policy between 1971 and 2007.

AB - This paper investigates the usefulness of information criteria for inference on the number of structural breaks in a standard linear regression model. In particular, we propose a modified penalty function for such criteria, which implies each break is equivalent to estimation of three individual regression coefficients. A Monte Carlo analysis compares information criteria to sequential testing, with the modified Bayesian and Hannan-Quinn criteria performing well overall, for data-generating processes both without and with breaks. The methods are also used to examine changes in Euro area monetary policy between 1971 and 2007.

UR - http://www.scopus.com/inward/record.url?scp=84880690471&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1111/manc.12017

U2 - 10.1111/manc.12017

DO - 10.1111/manc.12017

M3 - Article

VL - 81

SP - 54

EP - 81

JO - The Manchester School

JF - The Manchester School

SN - 1467-9957

IS - S3

ER -