Trend locally stationary wavelet processes
Trend locally stationary wavelet processes
Most time series observed in practice exhibit first- as well as second-order non-stationarity. In this article we propose a novel framework for modelling series with simultaneous time-varying first- and second-order structure, removing the restrictive zero-mean assumption of locally stationary wavelet processes and extending the applicability of the locally stationary wavelet model to include trend components. We develop an associated estimation theory for both first- and second-order time series quantities and show that our estimators achieve good properties in isolation of each other by making appropriate assumptions on the series trend. We demonstrate the utility of the method by analysing the global mean sea temperature time series, highlighting the impact of the changing climate.
895-917
McGonigle, Euan T.
1eec7a96-1343-4bf5-a131-432fe50842cd
Killick, Rebecca
c954436d-0b66-4ceb-bc63-bdf247ebee48
Nunes, Matthew A.
77bbd810-4dd8-4962-a2b3-c08205b4f54a
1 November 2022
McGonigle, Euan T.
1eec7a96-1343-4bf5-a131-432fe50842cd
Killick, Rebecca
c954436d-0b66-4ceb-bc63-bdf247ebee48
Nunes, Matthew A.
77bbd810-4dd8-4962-a2b3-c08205b4f54a
McGonigle, Euan T., Killick, Rebecca and Nunes, Matthew A.
(2022)
Trend locally stationary wavelet processes.
Journal of Time Series Analysis, 43 (6), .
(doi:10.1111/jtsa.12643).
Abstract
Most time series observed in practice exhibit first- as well as second-order non-stationarity. In this article we propose a novel framework for modelling series with simultaneous time-varying first- and second-order structure, removing the restrictive zero-mean assumption of locally stationary wavelet processes and extending the applicability of the locally stationary wavelet model to include trend components. We develop an associated estimation theory for both first- and second-order time series quantities and show that our estimators achieve good properties in isolation of each other by making appropriate assumptions on the series trend. We demonstrate the utility of the method by analysing the global mean sea temperature time series, highlighting the impact of the changing climate.
Text
Journal Time Series Analysis - 2022 - McGonigle - Trend locally stationary wavelet processes
- Version of Record
More information
Accepted/In Press date: 2 February 2022
e-pub ahead of print date: 8 February 2022
Published date: 1 November 2022
Identifiers
Local EPrints ID: 478216
URI: http://eprints.soton.ac.uk/id/eprint/478216
ISSN: 0143-9782
PURE UUID: 2d7760ae-8760-4529-9808-c9098348665d
Catalogue record
Date deposited: 23 Jun 2023 17:06
Last modified: 17 Mar 2024 04:20
Export record
Altmetrics
Contributors
Author:
Euan T. McGonigle
Author:
Rebecca Killick
Author:
Matthew A. Nunes
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics