Premium calculation using marginal generalized linear model combined with copula

Abstract

In order to determine non-life insurance premiums, we estimate the aggregate loss of insurance. Usually, in obtaining the aggregate loss model, we assume that the claim sizes and number of claims are independent. However, in some cases, there is a dependency between claim sizes and number of claims. We find the joint distribution of claim sizes and number of claims using copula. In this paper, we use the marginal Generalized Linear Model (GLM) that combined with a copula. The parameters of the marginal GLM will be estimated by maximum likelihood estimation. The estimate of aggregate loss is then used to find the non-life insurance premium.