Multiple change point detection and validation in autoregressive time series data

Lijing Ma; Andrew J. Grant; Georgy Sofronov

doi:10.1007/s00362-020-01198-w

Multiple change point detection and validation in autoregressive time series data

Lijing Ma, Andrew J. Grant, Georgy Sofronov

Research output: Contribution to journal › Conference paper › peer-review

15 Citations (Scopus)

74 Downloads (Pure)

Abstract

It is quite common that the structure of a time series changes abruptly. Identifying these change points and describing the model structure in the segments between these change points is of interest. In this paper, time series data is modelled assuming each segment is an autoregressive time series with possibly different autoregressive parameters. This is achieved using two main steps. The first step is to use a likelihood ratio scan based estimation technique to identify these potential change points to segment the time series. Once these potential change points are identified, modified parametric spectral discrimination tests are used to validate the proposed segments. A numerical study is conducted to demonstrate the performance of the proposed method across various scenarios and compared against other contemporary techniques.

Original language	English
Pages (from-to)	1507-1528
Number of pages	22
Journal	Statistical Papers
Volume	61
Issue number	4
Early online date	13 Jul 2020
DOIs	https://doi.org/10.1007/s00362-020-01198-w
Publication status	Published - Aug 2020
Event	Workshop on Change Point Detection Limit Theorems, Algorithms, and Applications in Life Sciences - Greifswald, Germany Duration: 8 Jul 2019 → 10 Jul 2019

Bibliographical note

Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

Changepoint detection
Autoregressive time series
Likelihood ratio scan statistics
Multiple testing problems

Access to Document

10.1007/s00362-020-01198-wLicence: CC BY

Publisher version (open access)Final published version, 0.98 MBLicence: CC BY

Cite this

@article{0f348cf2104548f4ad335891487c736d,

title = "Multiple change point detection and validation in autoregressive time series data",

abstract = "It is quite common that the structure of a time series changes abruptly. Identifying these change points and describing the model structure in the segments between these change points is of interest. In this paper, time series data is modelled assuming each segment is an autoregressive time series with possibly different autoregressive parameters. This is achieved using two main steps. The first step is to use a likelihood ratio scan based estimation technique to identify these potential change points to segment the time series. Once these potential change points are identified, modified parametric spectral discrimination tests are used to validate the proposed segments. A numerical study is conducted to demonstrate the performance of the proposed method across various scenarios and compared against other contemporary techniques.",

keywords = "Changepoint detection, Autoregressive time series, Likelihood ratio scan statistics, Multiple testing problems",

author = "Lijing Ma and Grant, {Andrew J.} and Georgy Sofronov",

note = "Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.; Workshop on Change Point Detection Limit Theorems, Algorithms, and Applications in Life Sciences ; Conference date: 08-07-2019 Through 10-07-2019",

year = "2020",

month = aug,

doi = "10.1007/s00362-020-01198-w",

language = "English",

volume = "61",

pages = "1507--1528",

journal = "Statistical Papers",

issn = "0932-5026",

publisher = "Springer, Springer Nature",

number = "4",

}

TY - JOUR

T1 - Multiple change point detection and validation in autoregressive time series data

AU - Ma, Lijing

AU - Grant, Andrew J.

AU - Sofronov, Georgy

N1 - Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2020/8

Y1 - 2020/8

N2 - It is quite common that the structure of a time series changes abruptly. Identifying these change points and describing the model structure in the segments between these change points is of interest. In this paper, time series data is modelled assuming each segment is an autoregressive time series with possibly different autoregressive parameters. This is achieved using two main steps. The first step is to use a likelihood ratio scan based estimation technique to identify these potential change points to segment the time series. Once these potential change points are identified, modified parametric spectral discrimination tests are used to validate the proposed segments. A numerical study is conducted to demonstrate the performance of the proposed method across various scenarios and compared against other contemporary techniques.

AB - It is quite common that the structure of a time series changes abruptly. Identifying these change points and describing the model structure in the segments between these change points is of interest. In this paper, time series data is modelled assuming each segment is an autoregressive time series with possibly different autoregressive parameters. This is achieved using two main steps. The first step is to use a likelihood ratio scan based estimation technique to identify these potential change points to segment the time series. Once these potential change points are identified, modified parametric spectral discrimination tests are used to validate the proposed segments. A numerical study is conducted to demonstrate the performance of the proposed method across various scenarios and compared against other contemporary techniques.

KW - Changepoint detection

KW - Autoregressive time series

KW - Likelihood ratio scan statistics

KW - Multiple testing problems

UR - http://www.scopus.com/inward/record.url?scp=85087766954&partnerID=8YFLogxK

U2 - 10.1007/s00362-020-01198-w

DO - 10.1007/s00362-020-01198-w

M3 - Conference paper

C2 - 33564212

AN - SCOPUS:85087766954

SN - 0932-5026

VL - 61

SP - 1507

EP - 1528

JO - Statistical Papers

JF - Statistical Papers

IS - 4

T2 - Workshop on Change Point Detection Limit Theorems, Algorithms, and Applications in Life Sciences

Y2 - 8 July 2019 through 10 July 2019

ER -

Multiple change point detection and validation in autoregressive time series data

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this