Home > Research > Publications & Outputs > Extreme events of Markov Chains

Electronic data

  • 16023_final

    Accepted author manuscript, 508 KB, PDF document

    Available under license: CC BY-NC

Links

Text available via DOI:

View graph of relations

Extreme events of Markov Chains

Research output: Contribution to journalJournal article

Published

Standard

Extreme events of Markov Chains. / Papastathopoulos, Ioannis; Strokorb, Kirstin; Tawn, Jonathan Angus; Butler, Adam.

In: Advances in Applied Probability, Vol. 49, No. 1, 03.2017, p. 134-161.

Research output: Contribution to journalJournal article

Harvard

Papastathopoulos, I, Strokorb, K, Tawn, JA & Butler, A 2017, 'Extreme events of Markov Chains', Advances in Applied Probability, vol. 49, no. 1, pp. 134-161. https://doi.org/10.1017/apr.2016.82

APA

Papastathopoulos, I., Strokorb, K., Tawn, J. A., & Butler, A. (2017). Extreme events of Markov Chains. Advances in Applied Probability, 49(1), 134-161. https://doi.org/10.1017/apr.2016.82

Vancouver

Papastathopoulos I, Strokorb K, Tawn JA, Butler A. Extreme events of Markov Chains. Advances in Applied Probability. 2017 Mar;49(1):134-161. https://doi.org/10.1017/apr.2016.82

Author

Papastathopoulos, Ioannis ; Strokorb, Kirstin ; Tawn, Jonathan Angus ; Butler, Adam. / Extreme events of Markov Chains. In: Advances in Applied Probability. 2017 ; Vol. 49, No. 1. pp. 134-161.

Bibtex

@article{58f02e7d45c84470aea54a0e45571630,
title = "Extreme events of Markov Chains",
abstract = "The extremal behaviour of a Markov chain is typically characterized by its tail chain. For asymptotically dependent Markov chains existing formulations fail to capture the full evolution of the extreme event when the chain moves out of the extreme tail region and for asymptotically independent chains recent results fail to cover well-known asymptotically independent processes such as Markov processes with a Gaussian copula between consecutive values. We use more sophisticated limiting mechanisms that cover a broader class of asymptotically independent processes than current methods, including an extension of the canonical Heffernan-Tawn normalization scheme, and reveal features which existing methods reduce to a degenerate form associated with non-extreme states.",
author = "Ioannis Papastathopoulos and Kirstin Strokorb and Tawn, {Jonathan Angus} and Adam Butler",
year = "2017",
month = "3",
doi = "10.1017/apr.2016.82",
language = "English",
volume = "49",
pages = "134--161",
journal = "Advances in Applied Probability",
issn = "0001-8678",
publisher = "Cambridge University Press",
number = "1",

}

RIS

TY - JOUR

T1 - Extreme events of Markov Chains

AU - Papastathopoulos, Ioannis

AU - Strokorb, Kirstin

AU - Tawn, Jonathan Angus

AU - Butler, Adam

PY - 2017/3

Y1 - 2017/3

N2 - The extremal behaviour of a Markov chain is typically characterized by its tail chain. For asymptotically dependent Markov chains existing formulations fail to capture the full evolution of the extreme event when the chain moves out of the extreme tail region and for asymptotically independent chains recent results fail to cover well-known asymptotically independent processes such as Markov processes with a Gaussian copula between consecutive values. We use more sophisticated limiting mechanisms that cover a broader class of asymptotically independent processes than current methods, including an extension of the canonical Heffernan-Tawn normalization scheme, and reveal features which existing methods reduce to a degenerate form associated with non-extreme states.

AB - The extremal behaviour of a Markov chain is typically characterized by its tail chain. For asymptotically dependent Markov chains existing formulations fail to capture the full evolution of the extreme event when the chain moves out of the extreme tail region and for asymptotically independent chains recent results fail to cover well-known asymptotically independent processes such as Markov processes with a Gaussian copula between consecutive values. We use more sophisticated limiting mechanisms that cover a broader class of asymptotically independent processes than current methods, including an extension of the canonical Heffernan-Tawn normalization scheme, and reveal features which existing methods reduce to a degenerate form associated with non-extreme states.

U2 - 10.1017/apr.2016.82

DO - 10.1017/apr.2016.82

M3 - Journal article

VL - 49

SP - 134

EP - 161

JO - Advances in Applied Probability

JF - Advances in Applied Probability

SN - 0001-8678

IS - 1

ER -