Home > Research > Publications & Outputs > Most recent changepoint detection in Panel data

Electronic data

  • draft_MRC

    Accepted author manuscript, 332 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Most recent changepoint detection in Panel data

Research output: Contribution to journalJournal article

Published

Standard

Most recent changepoint detection in Panel data. / Bardwell, Lawrence; Fearnhead, Paul Nicholas; Eckley, Idris Arthur; Smith, Simon; Spott, Martin.

In: Technometrics, Vol. 61, No. 1, 02.01.2019, p. 88-98.

Research output: Contribution to journalJournal article

Harvard

APA

Vancouver

Author

Bibtex

@article{96e699ec7fd14763b2c9496bdd35bd8b,
title = "Most recent changepoint detection in Panel data",
abstract = "Detecting recent changepoints in time-series can be important for short-term prediction, as we can then base predictions just on the data since the changepoint. In many applications we have panel data, consisting of many related univariate time-series. We present a novel approach to detect sets of most recent changepoints in such panel data which aims to pool information across time-series, so that we preferentially infer a most recent change at the same time-point in multiple series. Our approach is computationally efficient as it involves analysing each time-series independently to obtain a profile likelihood like quantity that summarises the evidence for the series having either no change or a specific value for its most recent changepoint. We then post-process this output from each time-series to obtain a potentially small set of times for the most recent changepoints, and, for each time, the set of series which has their most recent changepoint at that time. We demonstrate the usefulness of this method on two data sets: forecasting events in a telecommunications network and inference about changes in the net asset ratio for a panel of US firms.",
keywords = "Breakpoints, Changepoints, Forecasting, Panel Data, Structural Breaks",
author = "Lawrence Bardwell and Fearnhead, {Paul Nicholas} and Eckley, {Idris Arthur} and Simon Smith and Martin Spott",
year = "2019",
month = jan
day = "2",
doi = "10.1080/00401706.2018.1438926",
language = "English",
volume = "61",
pages = "88--98",
journal = "Technometrics",
issn = "0040-1706",
publisher = "American Statistical Association",
number = "1",

}

RIS

TY - JOUR

T1 - Most recent changepoint detection in Panel data

AU - Bardwell, Lawrence

AU - Fearnhead, Paul Nicholas

AU - Eckley, Idris Arthur

AU - Smith, Simon

AU - Spott, Martin

PY - 2019/1/2

Y1 - 2019/1/2

N2 - Detecting recent changepoints in time-series can be important for short-term prediction, as we can then base predictions just on the data since the changepoint. In many applications we have panel data, consisting of many related univariate time-series. We present a novel approach to detect sets of most recent changepoints in such panel data which aims to pool information across time-series, so that we preferentially infer a most recent change at the same time-point in multiple series. Our approach is computationally efficient as it involves analysing each time-series independently to obtain a profile likelihood like quantity that summarises the evidence for the series having either no change or a specific value for its most recent changepoint. We then post-process this output from each time-series to obtain a potentially small set of times for the most recent changepoints, and, for each time, the set of series which has their most recent changepoint at that time. We demonstrate the usefulness of this method on two data sets: forecasting events in a telecommunications network and inference about changes in the net asset ratio for a panel of US firms.

AB - Detecting recent changepoints in time-series can be important for short-term prediction, as we can then base predictions just on the data since the changepoint. In many applications we have panel data, consisting of many related univariate time-series. We present a novel approach to detect sets of most recent changepoints in such panel data which aims to pool information across time-series, so that we preferentially infer a most recent change at the same time-point in multiple series. Our approach is computationally efficient as it involves analysing each time-series independently to obtain a profile likelihood like quantity that summarises the evidence for the series having either no change or a specific value for its most recent changepoint. We then post-process this output from each time-series to obtain a potentially small set of times for the most recent changepoints, and, for each time, the set of series which has their most recent changepoint at that time. We demonstrate the usefulness of this method on two data sets: forecasting events in a telecommunications network and inference about changes in the net asset ratio for a panel of US firms.

KW - Breakpoints

KW - Changepoints

KW - Forecasting

KW - Panel Data

KW - Structural Breaks

U2 - 10.1080/00401706.2018.1438926

DO - 10.1080/00401706.2018.1438926

M3 - Journal article

VL - 61

SP - 88

EP - 98

JO - Technometrics

JF - Technometrics

SN - 0040-1706

IS - 1

ER -