Clustering seasonal time series using extreme value analysis: An application to Spanish temperature time series

Elizabeth A. Maharaj, Andres Modesto Alonso, Pierpaolo D'Urso

Research output: Contribution to journalArticleResearchpeer-review

Abstract

A challenging aspect of grouping together regional temperature time series is that some regions have similar summer temperatures but different winter temperatures and vice versa. We explore this by applying cluster analysis to regional temperature time series in Spain using as features the parameter estimates of location, scale, and shape, obtained from fitting the generalized extreme value (GEV) distribution to the block maxima and block minima of the series. Using this approach, our findings reveal that the identified clusters can be meaningfully interpreted and are well validated. The motivation for using this approach is that each time series is represented by just three easily extracted features. If features were to be extracted as a result of conventional time series modeling, they are likely to be impacted upon by the uncertainty of model selection. This is not the case with GEV modeling. Furthermore, GEV modeling enables long – term projections of the maxima and minima that cannot otherwise be achieved from conventional time series modeling. For comparison purposes, we also explore clustering the block maxima and block minima of the times series. In addition, we explore the performance of this approach using simulated data.
Original languageEnglish
Pages (from-to)175-191
Number of pages17
JournalCommunications in Statistics: Case Studies, Data Analysis and Applications
Volume1
Issue number4
DOIs
Publication statusPublished - 2016

Keywords

  • block maxima
  • clustering
  • generalized extreme value analysis
  • returns

Cite this

@article{a85d6ba489214a068172bbbb37ec9828,
title = "Clustering seasonal time series using extreme value analysis: An application to Spanish temperature time series",
abstract = "A challenging aspect of grouping together regional temperature time series is that some regions have similar summer temperatures but different winter temperatures and vice versa. We explore this by applying cluster analysis to regional temperature time series in Spain using as features the parameter estimates of location, scale, and shape, obtained from fitting the generalized extreme value (GEV) distribution to the block maxima and block minima of the series. Using this approach, our findings reveal that the identified clusters can be meaningfully interpreted and are well validated. The motivation for using this approach is that each time series is represented by just three easily extracted features. If features were to be extracted as a result of conventional time series modeling, they are likely to be impacted upon by the uncertainty of model selection. This is not the case with GEV modeling. Furthermore, GEV modeling enables long – term projections of the maxima and minima that cannot otherwise be achieved from conventional time series modeling. For comparison purposes, we also explore clustering the block maxima and block minima of the times series. In addition, we explore the performance of this approach using simulated data.",
keywords = "block maxima, clustering, generalized extreme value analysis, returns",
author = "Maharaj, {Elizabeth A.} and Alonso, {Andres Modesto} and Pierpaolo D'Urso",
year = "2016",
doi = "10.1080/23737484.2016.1179140",
language = "English",
volume = "1",
pages = "175--191",
journal = "Communications in Statistics: Case Studies, Data Analysis and Applications",
issn = "2373-7484",
publisher = "Taylor & Francis",
number = "4",

}

Clustering seasonal time series using extreme value analysis : An application to Spanish temperature time series. / Maharaj, Elizabeth A.; Alonso, Andres Modesto; D'Urso, Pierpaolo.

In: Communications in Statistics: Case Studies, Data Analysis and Applications, Vol. 1, No. 4, 2016, p. 175-191.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Clustering seasonal time series using extreme value analysis

T2 - An application to Spanish temperature time series

AU - Maharaj, Elizabeth A.

AU - Alonso, Andres Modesto

AU - D'Urso, Pierpaolo

PY - 2016

Y1 - 2016

N2 - A challenging aspect of grouping together regional temperature time series is that some regions have similar summer temperatures but different winter temperatures and vice versa. We explore this by applying cluster analysis to regional temperature time series in Spain using as features the parameter estimates of location, scale, and shape, obtained from fitting the generalized extreme value (GEV) distribution to the block maxima and block minima of the series. Using this approach, our findings reveal that the identified clusters can be meaningfully interpreted and are well validated. The motivation for using this approach is that each time series is represented by just three easily extracted features. If features were to be extracted as a result of conventional time series modeling, they are likely to be impacted upon by the uncertainty of model selection. This is not the case with GEV modeling. Furthermore, GEV modeling enables long – term projections of the maxima and minima that cannot otherwise be achieved from conventional time series modeling. For comparison purposes, we also explore clustering the block maxima and block minima of the times series. In addition, we explore the performance of this approach using simulated data.

AB - A challenging aspect of grouping together regional temperature time series is that some regions have similar summer temperatures but different winter temperatures and vice versa. We explore this by applying cluster analysis to regional temperature time series in Spain using as features the parameter estimates of location, scale, and shape, obtained from fitting the generalized extreme value (GEV) distribution to the block maxima and block minima of the series. Using this approach, our findings reveal that the identified clusters can be meaningfully interpreted and are well validated. The motivation for using this approach is that each time series is represented by just three easily extracted features. If features were to be extracted as a result of conventional time series modeling, they are likely to be impacted upon by the uncertainty of model selection. This is not the case with GEV modeling. Furthermore, GEV modeling enables long – term projections of the maxima and minima that cannot otherwise be achieved from conventional time series modeling. For comparison purposes, we also explore clustering the block maxima and block minima of the times series. In addition, we explore the performance of this approach using simulated data.

KW - block maxima

KW - clustering

KW - generalized extreme value analysis

KW - returns

U2 - 10.1080/23737484.2016.1179140

DO - 10.1080/23737484.2016.1179140

M3 - Article

VL - 1

SP - 175

EP - 191

JO - Communications in Statistics: Case Studies, Data Analysis and Applications

JF - Communications in Statistics: Case Studies, Data Analysis and Applications

SN - 2373-7484

IS - 4

ER -