Dependence modelling on loss triangles: copula regression with unobserved effect

Karatas R., KARABEY U.

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2025 (SCI-Expanded, Scopus) identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1080/03610918.2025.2496760
  • Dergi Adı: COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Business Source Elite, Business Source Premier, CAB Abstracts, Compendex, Computer & Applied Sciences, Veterinary Science Database, zbMATH, Civil Engineering Abstracts
  • Hacettepe Üniversitesi Adresli: Evet

Özet

Classical reserve calculation techniques assume that the loss amounts are independent. However, various dependency structures, such as accident year dependence, development year dependence, and calendar year dependence, exist. In this study, we assume the presence of an unobserved random effect both between and within lines of business. In order to model this effect, we propose a copula regression model with unobserved random effect in which we consider additive and multiplicative effect structure. We consider four different random effect distributions to model the unobserved random effect. The theoretical results are illustrated using a real loss triangle dataset for two dependent lines of business. Based on the results, we conclude that proposed model outperforms classical models.