Robust claim frequency modeling through phase-type mixture-of-experts regression

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Robust claim frequency modeling through phase-type mixture-of-experts regression. / Bladt, Martin; Yslas, Jorge.

In: Insurance: Mathematics and Economics, Vol. 111, 2023, p. 1-22.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Bladt, M & Yslas, J 2023, 'Robust claim frequency modeling through phase-type mixture-of-experts regression', Insurance: Mathematics and Economics, vol. 111, pp. 1-22. https://doi.org/10.1016/j.insmatheco.2023.02.008

APA

Bladt, M., & Yslas, J. (2023). Robust claim frequency modeling through phase-type mixture-of-experts regression. Insurance: Mathematics and Economics, 111, 1-22. https://doi.org/10.1016/j.insmatheco.2023.02.008

Vancouver

Bladt M, Yslas J. Robust claim frequency modeling through phase-type mixture-of-experts regression. Insurance: Mathematics and Economics. 2023;111:1-22. https://doi.org/10.1016/j.insmatheco.2023.02.008

Author

Bladt, Martin ; Yslas, Jorge. / Robust claim frequency modeling through phase-type mixture-of-experts regression. In: Insurance: Mathematics and Economics. 2023 ; Vol. 111. pp. 1-22.

Bibtex

@article{f17fc5e5b6264633a6834fb52a50c315,
title = "Robust claim frequency modeling through phase-type mixture-of-experts regression",
abstract = "This paper addresses the problem of modeling loss frequency using regression when the counts have a non-standard distribution. We propose a novel approach based on mixture-of-experts specifications on discrete-phase type distributions. Compared to continuous phase-type counterparts, our approach offers fast estimation via expectation-maximization, making it more feasible for use in real-life scenarios. Our model is both robust and interpretable in terms of risk classes, and can be naturally extended to the multivariate case through two different constructions. This avoids the need for ad-hoc multivariate claim count modeling. Overall, our approach provides a more effective solution for modeling loss frequency in non-standard situations.",
keywords = "Claim count distributions, Discrete phase-type distributions, Regression modeling",
author = "Martin Bladt and Jorge Yslas",
note = "Publisher Copyright: {\textcopyright} 2023 The Author(s)",
year = "2023",
doi = "10.1016/j.insmatheco.2023.02.008",
language = "English",
volume = "111",
pages = "1--22",
journal = "Insurance: Mathematics and Economics",
issn = "0167-6687",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - Robust claim frequency modeling through phase-type mixture-of-experts regression

AU - Bladt, Martin

AU - Yslas, Jorge

N1 - Publisher Copyright: © 2023 The Author(s)

PY - 2023

Y1 - 2023

N2 - This paper addresses the problem of modeling loss frequency using regression when the counts have a non-standard distribution. We propose a novel approach based on mixture-of-experts specifications on discrete-phase type distributions. Compared to continuous phase-type counterparts, our approach offers fast estimation via expectation-maximization, making it more feasible for use in real-life scenarios. Our model is both robust and interpretable in terms of risk classes, and can be naturally extended to the multivariate case through two different constructions. This avoids the need for ad-hoc multivariate claim count modeling. Overall, our approach provides a more effective solution for modeling loss frequency in non-standard situations.

AB - This paper addresses the problem of modeling loss frequency using regression when the counts have a non-standard distribution. We propose a novel approach based on mixture-of-experts specifications on discrete-phase type distributions. Compared to continuous phase-type counterparts, our approach offers fast estimation via expectation-maximization, making it more feasible for use in real-life scenarios. Our model is both robust and interpretable in terms of risk classes, and can be naturally extended to the multivariate case through two different constructions. This avoids the need for ad-hoc multivariate claim count modeling. Overall, our approach provides a more effective solution for modeling loss frequency in non-standard situations.

KW - Claim count distributions

KW - Discrete phase-type distributions

KW - Regression modeling

UR - http://www.scopus.com/inward/record.url?scp=85149902744&partnerID=8YFLogxK

U2 - 10.1016/j.insmatheco.2023.02.008

DO - 10.1016/j.insmatheco.2023.02.008

M3 - Journal article

AN - SCOPUS:85149902744

VL - 111

SP - 1

EP - 22

JO - Insurance: Mathematics and Economics

JF - Insurance: Mathematics and Economics

SN - 0167-6687

ER -

ID: 359611679