Confidence sets for the date of a break in level and trend when the order of integration is unknown

We propose methods for constructing confidence sets for the timing of a break in level and/or trend that have asymptotically correct coverage for both I(0) and I(1) processes. These are based on inverting a sequence of tests for the break location, evaluated across all possible break dates. We separ...

Full description

Bibliographic Details
Main Authors: Harvey, David I., Leybourne, Stephen J.
Format: Article
Published: Elsevier 2015
Subjects:
Online Access:https://eprints.nottingham.ac.uk/32662/
_version_ 1848794461742563328
author Harvey, David I.
Leybourne, Stephen J.
author_facet Harvey, David I.
Leybourne, Stephen J.
author_sort Harvey, David I.
building Nottingham Research Data Repository
collection Online Access
description We propose methods for constructing confidence sets for the timing of a break in level and/or trend that have asymptotically correct coverage for both I(0) and I(1) processes. These are based on inverting a sequence of tests for the break location, evaluated across all possible break dates. We separately derive locally best invariant tests for the I(0) and I(1) cases; under their respective assumptions, the resulting confidence sets provide correct asymptotic coverage regardless of the magnitude of the break. We suggest use of a pre-test procedure to select between the I(0)- and I(1)-based confidence sets, and Monte Carlo evidence demonstrates that our recommended procedure achieves good finite sample properties in terms of coverage and length across both I(0) and I(1) environments. An application using US macroeconomic data is provided which further evinces the value of these procedures.
first_indexed 2025-11-14T19:16:34Z
format Article
id nottingham-32662
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:16:34Z
publishDate 2015
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-326622020-05-04T20:10:03Z https://eprints.nottingham.ac.uk/32662/ Confidence sets for the date of a break in level and trend when the order of integration is unknown Harvey, David I. Leybourne, Stephen J. We propose methods for constructing confidence sets for the timing of a break in level and/or trend that have asymptotically correct coverage for both I(0) and I(1) processes. These are based on inverting a sequence of tests for the break location, evaluated across all possible break dates. We separately derive locally best invariant tests for the I(0) and I(1) cases; under their respective assumptions, the resulting confidence sets provide correct asymptotic coverage regardless of the magnitude of the break. We suggest use of a pre-test procedure to select between the I(0)- and I(1)-based confidence sets, and Monte Carlo evidence demonstrates that our recommended procedure achieves good finite sample properties in terms of coverage and length across both I(0) and I(1) environments. An application using US macroeconomic data is provided which further evinces the value of these procedures. Elsevier 2015-02 Article PeerReviewed Harvey, David I. and Leybourne, Stephen J. (2015) Confidence sets for the date of a break in level and trend when the order of integration is unknown. Journal of Econometrics, 184 (2). pp. 262-279. ISSN 0304-4076 Level break; Trend break; Stationary; Unit root; Locally best invariant test; Confidence sets http://www.sciencedirect.com/science/article/pii/S0304407614001894 doi:10.1016/j.jeconom.2014.09.004 doi:10.1016/j.jeconom.2014.09.004
spellingShingle Level break; Trend break; Stationary; Unit root; Locally best invariant test; Confidence sets
Harvey, David I.
Leybourne, Stephen J.
Confidence sets for the date of a break in level and trend when the order of integration is unknown
title Confidence sets for the date of a break in level and trend when the order of integration is unknown
title_full Confidence sets for the date of a break in level and trend when the order of integration is unknown
title_fullStr Confidence sets for the date of a break in level and trend when the order of integration is unknown
title_full_unstemmed Confidence sets for the date of a break in level and trend when the order of integration is unknown
title_short Confidence sets for the date of a break in level and trend when the order of integration is unknown
title_sort confidence sets for the date of a break in level and trend when the order of integration is unknown
topic Level break; Trend break; Stationary; Unit root; Locally best invariant test; Confidence sets
url https://eprints.nottingham.ac.uk/32662/
https://eprints.nottingham.ac.uk/32662/
https://eprints.nottingham.ac.uk/32662/