Home > Research > Publications & Outputs > Degree distributions in networks:

Electronic data

  • 2008.03073v4

    Accepted author manuscript, 2.68 MB, PDF document

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

Links

Text available via DOI:

View graph of relations

Degree distributions in networks: beyond the power law

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Degree distributions in networks: beyond the power law. / Lee, Clement; Eastoe, Emma; Farrell, Aiden.
In: Statistica Neerlandica, Vol. 78, No. 4, 30.11.2024, p. 702-718.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Lee C, Eastoe E, Farrell A. Degree distributions in networks: beyond the power law. Statistica Neerlandica. 2024 Nov 30;78(4):702-718. Epub 2024 Jul 23. doi: 10.1111/stan.12355

Author

Lee, Clement ; Eastoe, Emma ; Farrell, Aiden. / Degree distributions in networks: beyond the power law. In: Statistica Neerlandica. 2024 ; Vol. 78, No. 4. pp. 702-718.

Bibtex

@article{a8150f3db8b94eeb896a216b6100b74c,
title = "Degree distributions in networks:: beyond the power law",
abstract = "The power law is useful in describing count phenomena such as network degreesand word frequencies. With a single parameter, it captures the main feature that the frequencies are linear on the log-log scale. Nevertheless, there have been criticisms of the power law, for example that a threshold needs to be pre-selected without its uncertainty quantified, that the power law is simply inadequate, and that subsequent hypothesis tests are required to determine whether the data could have come from the power law. We propose a modelling framework that combines two different generalisations of the power law, namely the generalised Pareto distribution and the Zipf-polylog distribution, to resolve these issues. The proposed mixture distributions are shown to fit the data well and quantify the threshold uncertainty in a natural way. A model selection step embedded in the Bayesian inference algorithm further answers the question whether the power law is adequate.",
author = "Clement Lee and Emma Eastoe and Aiden Farrell",
year = "2024",
month = nov,
day = "30",
doi = "10.1111/stan.12355",
language = "English",
volume = "78",
pages = "702--718",
journal = "Statistica Neerlandica",
issn = "0039-0402",
publisher = "Wiley-Blackwell",
number = "4",

}

RIS

TY - JOUR

T1 - Degree distributions in networks:

T2 - beyond the power law

AU - Lee, Clement

AU - Eastoe, Emma

AU - Farrell, Aiden

PY - 2024/11/30

Y1 - 2024/11/30

N2 - The power law is useful in describing count phenomena such as network degreesand word frequencies. With a single parameter, it captures the main feature that the frequencies are linear on the log-log scale. Nevertheless, there have been criticisms of the power law, for example that a threshold needs to be pre-selected without its uncertainty quantified, that the power law is simply inadequate, and that subsequent hypothesis tests are required to determine whether the data could have come from the power law. We propose a modelling framework that combines two different generalisations of the power law, namely the generalised Pareto distribution and the Zipf-polylog distribution, to resolve these issues. The proposed mixture distributions are shown to fit the data well and quantify the threshold uncertainty in a natural way. A model selection step embedded in the Bayesian inference algorithm further answers the question whether the power law is adequate.

AB - The power law is useful in describing count phenomena such as network degreesand word frequencies. With a single parameter, it captures the main feature that the frequencies are linear on the log-log scale. Nevertheless, there have been criticisms of the power law, for example that a threshold needs to be pre-selected without its uncertainty quantified, that the power law is simply inadequate, and that subsequent hypothesis tests are required to determine whether the data could have come from the power law. We propose a modelling framework that combines two different generalisations of the power law, namely the generalised Pareto distribution and the Zipf-polylog distribution, to resolve these issues. The proposed mixture distributions are shown to fit the data well and quantify the threshold uncertainty in a natural way. A model selection step embedded in the Bayesian inference algorithm further answers the question whether the power law is adequate.

UR - https://doi.org/10.48550/arXiv.2008.03073

U2 - 10.1111/stan.12355

DO - 10.1111/stan.12355

M3 - Journal article

VL - 78

SP - 702

EP - 718

JO - Statistica Neerlandica

JF - Statistica Neerlandica

SN - 0039-0402

IS - 4

ER -