Paper Title
Estimation Frequency and Severity of Claims on Health Insurance using the Copula-Based Regression Model (Case Study of PT B Insurance)

Abstract
This paperanalyzes the frequency and severity of claims which are the two main risks in general insurance. Determination of insurance contract prices from policyholders by multiplying expected frequency of claims and severity of claims. The Net Premium is an estimate of the aggregate loss of a policy group that models it assuming the frequency and severity of the claims are independent. However, in some cases, there are dependencies between the two variables. To overcome this dependency problem, this thesis uses copula – based regression model to build a joint distribution. This is by combining the marginal generalized linear model of frequency and the average severity of claims using copula. The parameters of the distribution are estimated using the maximum likelihood method. Further, the selection of the best copula that will be used in building a joint distribution is done by looking at the greatest log-likelihood value and smallest value of Root Mean Square Error (RMSE). The result is clayton copula based regression model is chosen. Finally, estimatie of the frequency and average severity of claims is calculated and policy losses are generated based on the estimated mean value of the joint distribution. Keywords - General Insurance, Claim Frequency, Claim Severity Generalized Linear Model, Copula, Policy Loss.