Multi-Peril Frequency Credibility Premium via Shared Random Effects
Variance · 08.01.2024 · 17 (1)
DOI: 10.66573/001c.91290 ↗Kurz gefasst
Fachartikel in der Zeitschrift Variance, online erschienen am 08.01.2024. Verfasst von Himchan Jeong und 1 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben.
Abstract (Originalsprache)
It is not uncommon to observe that an insurance policy consists of multiple coverages that cover different perils over multiple years. In this article, we propose a shared random effects model to not only capture the unobserved heterogeneity in risks but also induce a natural dependence structure among the claims from multiple perils. The proposed model has a nice interpretation and comes with closed forms of joint likelihood and credibility premiums that are desirable for ease of implementation in practice. Both the proposed method and benchmark methods are calibrated using a public insurance dataset provided by the Wisconsin Local Government Property Insurance Fund, where one can observe both policy characteristics and historical claims information. It turns out that the proposed method shows acceptable level of performance compared to many pre-existing benchmarks in terms of out-of-sample validation. Address for Correspondence: himchan_jeong@sfu.ca
Zitieren
APA
Jeong, H., & Dey, D. K. (2024). Multi-Peril Frequency Credibility Premium via Shared Random Effects. Variance, 17(1). https://doi.org/10.66573/001c.91290
Vancouver
Jeong H, Dey DK. Multi-Peril Frequency Credibility Premium via Shared Random Effects. Variance. 2024;17(1). doi:10.66573/001c.91290
BibTeX
@article{Jeong2024,
author = {Jeong, Himchan and Dey, Dipak Kumar},
title = {Multi-Peril Frequency Credibility Premium via Shared Random Effects},
journal = {Variance},
year = {2024},
volume = {17},
number = {1},
doi = {10.66573/001c.91290}
}
EUVIMED