One method for calculating premiums based on the policyholder's risk characteristics is to employ the theory of credibility, particularly the semiparametric Bühlmann model. The aim of this research is to estimate the parameters of the Bühlmann credibility model using a semiparametric approach for claim frequencies that follow a Poisson distribution. Additionally, this study compares the semiparametric model, the parametric model, and the nonparametric model for the Bühlmann model. The assumptions made in this study concern claim frequencies following a Poisson distribution. The research results reveal that the semiparametric Bühlmann credibility premium for a Poisson distribution is 0.117992. Furthermore, the comparison between parametric and semiparametric approaches shows that premiums estimated using the semiparametric approach are lower than those estimated using the parametric approach. The difference is approximately 0.0085% for the Negative Binomial distribution and 0.00085% for the two Poisson distributions. However, there is no significant difference in premium values between the semiparametric and nonparametric approaches.
Copyrights © 2023