Credibility Theory for Variance Premium Principle
In this paper, the authors introduce a credibility approach to estimate a linear combination of hypothetical mean and process variance by minimizing the expected quadratic loss function. The estimator they introduced consists of the linear form of observations, their quadratic terms and some quantities containing population information. The authors compare a spin-off result with the classical model and the \(q\)-credibility model. They also provide nonparametric estimators of structural quantities and numerical results to illustrate the performance of the estimators. They use a real insurance dataset to illustrate a data-driven procedure for determining the safety loading. In Section 2 of the paper, the authors review some results in the classical credibility model and the \(q\)-credibility model. Then they derive a credibility estimator for the variance premium principle by minimizing the expected quadratic loss function. They analyse some properties of the mean squared errors of the proposed estimator. The main result of the analysis is presented in Theorem 1. Numerical examples illustrating the performance of the estimator are provided based on discrete and continuous distributions with both light-tailed and heavy-tailed distributions. A nonparametric study for the structural parameters is provided in Section 3. Both simulation data and real insurance data are used for the illustrations. The authors also introduce a data-dependent method to determine the number of groups of risk profiles. The main theoretical results for the nonparametric estimators are presented in Proposition 3, Proposition 4, and Proposition 5. The final section (Section 4) gives the conclusion.
- A Bayesian nonparametric model and its application in insurance loss prediction
- A course in credibility theory and its applications
- A flexible Bayesian nonparametric model for predicting future insurance claims prediction
- A general optimal approach to Bühlmann credibility theory
- A new approach to the credibility formula
- Bayesian nonparametric predictive modeling of group health claims
- Bayesian nonparametric regression models for modeling and predicting healthcare claims
- Credibility theory: A new view from the theory of second order optimal statistics.
- Evolutionary credibility risk premium
- Evolutionary Credibility Theory
- Exact credibility reference Bayesian premiums
- scientific article; zbMATH DE number 3551731 (Why is no real title available?)
- Mean-variance insurance design with counterparty risk and incentive compatibility
- Moral-hazard-free insurance: mean-variance premium principle and rank-dependent utility theory
- Optimal reinsurance under mean-variance premium principles
- Pure robust versus robust portfolio unbiased -- credibility and asymptotic optimality
- Quantile credibility models
- Satisficing credibility for heterogeneous risks
This page was built for publication: Credibility Theory for Variance Premium Principle
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7000901)
