Testing for structural breaks in the presence of data perturbations: impacts and wavelet-based improvements
(2015) In Journal of Statistical Computation and Simulation 85(17). p.3468-3479- Abstract
- This paper investigates how classical measurement error and additive outliers (AO) influence tests for structural change based on F-statistics. We derive theoretically the impact of general additive disturbances in the regressors on the asymptotic distribution of these tests for structural change. The small sample properties in the case of classical measurement error and AO are investigated via Monte Carlo simulations, revealing that sizes are biased upwards and that powers are reduced. Two-wavelet-based denoising methods are used to reduce these distortions. We show that these two methods can significantly improve the performance of structural break tests.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/4905262
- author
- Reese, Simon LU and Li, Yushu
- organization
- publishing date
- 2015
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- structural breaks, measurement error, additive outlier, wavelet transform, empirical Bayes thresholding
- in
- Journal of Statistical Computation and Simulation
- volume
- 85
- issue
- 17
- pages
- 3468 - 3479
- publisher
- Taylor & Francis
- external identifiers
-
- scopus:84941743297
- wos:000371315300005
- ISSN
- 1563-5163
- DOI
- 10.1080/00949655.2014.979824
- language
- English
- LU publication?
- yes
- id
- 6196d274-ff93-4cfa-8af0-82d5488f7478 (old id 4905262)
- alternative location
- http://www.tandfonline.com/doi/full/10.1080/00949655.2014.979824
- date added to LUP
- 2016-04-01 09:54:53
- date last changed
- 2022-01-25 17:52:55
@article{6196d274-ff93-4cfa-8af0-82d5488f7478,
abstract = {{This paper investigates how classical measurement error and additive outliers (AO) influence tests for structural change based on F-statistics. We derive theoretically the impact of general additive disturbances in the regressors on the asymptotic distribution of these tests for structural change. The small sample properties in the case of classical measurement error and AO are investigated via Monte Carlo simulations, revealing that sizes are biased upwards and that powers are reduced. Two-wavelet-based denoising methods are used to reduce these distortions. We show that these two methods can significantly improve the performance of structural break tests.}},
author = {{Reese, Simon and Li, Yushu}},
issn = {{1563-5163}},
keywords = {{structural breaks; measurement error; additive outlier; wavelet transform; empirical Bayes thresholding}},
language = {{eng}},
number = {{17}},
pages = {{3468--3479}},
publisher = {{Taylor & Francis}},
series = {{Journal of Statistical Computation and Simulation}},
title = {{Testing for structural breaks in the presence of data perturbations: impacts and wavelet-based improvements}},
url = {{http://dx.doi.org/10.1080/00949655.2014.979824}},
doi = {{10.1080/00949655.2014.979824}},
volume = {{85}},
year = {{2015}},
}