Taryfikacja a priori z wykorzystaniem kopuli
On the use of copula in ratemaking

Summary/Abstract: In the current insurance practice in ratemaking the generalized linear models GLM are used. Most often, separately expected value of expected claim frequency and expected value of claim severity for individual risk taking important thesis of independence of both variables are estimated. The article confronts an alternative approach in which a relationship between these modeled variables using a copula function is acceptable. In the proposed model there were estimated both variables distribution joint as well as their marginal distributions. This allowed the estimation of the expected value of the total claim amount which is interpreted as a pure risk premium for individual risk. For the estimation there was used the likelihood function containing a specified copula function. To illustrate the functioning of the model a numerical example was shown/presented/demonstrated, which uses data from motor insurance databases. The calculations were carried out using the {CopulaRegression} R-package.

• Details
• Contents