Count copula regression model using generalized beta distribution of the second kind

Hadi Safari-Katesari, Samira Zaroudi

Research output: Contribution to journalArticlepeer-review

Abstract

Modelling claims severity for obtaining insurance premium is one of the major concerns of the insurance industry. There is a considerable amount of literature on the actuarial application of the copula model to calculate the pure premium. In this paper, we model claims severity for computing the pure premium in the collision market by means of the count copula model. Moreover, we apply a regression model using a generalized beta distribution of the second kind (GB2) to compute the premium for an average claim and the conditional computation for all coverage levels. Like many other researchers, we assume that the number of accidents is independent from the size of claims. For real data application, we use a portfolio of a major automobile insurer in Iran in 2007-2008, with a subsample of 59,547 policies available in their portfolio. We then proceed to compare the estimated premiums with the real premiums. The results demonstrate that there is strong positive dependency between the real premium and the estimated one.

Original languageEnglish
Pages (from-to)1-12
Number of pages12
JournalStatistics in Transition New Series
Volume21
Issue number2
DOIs
StatePublished - Jun 2020

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'Count copula regression model using generalized beta distribution of the second kind'. Together they form a unique fingerprint.

Cite this