On Stochastic Approximation and Credibility
Let \(X\) be a claim size with distribution \(P_{\theta}\), where \({\theta}\) denotes the risk parameter with prior distribution \(\pi(\theta)\). The problem of estimating the fair premium \(\mu=E(X | \theta)\), given \(n\) years individual experience \(x_1,x_2,\dots,x_n\), and the collective fair premium \(m=\int \mu(\theta)\pi(\theta)d\theta)\), can be solved by means of the credibility formula \(\mu_n=(1-\alpha_n)m+\alpha_n \bar x_n\), where \(a_n\) is a reliability factor. The author used such powerful tools for sequential estimation as stochastic approximation techniques to obtain certain analogues of the credibility formula for families of distributions \(\{P_{\theta}\}\) for which no simple credibility formula exists. For example, the so-called exponential dispersion family (EDF) and location dispersion family (LDF) are considered and an optimal stepwise gain sequence for LDF is obtained.
- A Stochastic Approximation Method
- Conjugate priors for exponential families
- Exponential dispersion models and credibility
- Exponential Dispersion Models and Extensions: A Review
- Exponential transformation models
- Fitting Tweedie's compound poisson model to insurance claims data
- scientific article; zbMATH DE number 3426675 (Why is no real title available?)
- scientific article; zbMATH DE number 3723610 (Why is no real title available?)
- scientific article; zbMATH DE number 1128584 (Why is no real title available?)
- scientific article; zbMATH DE number 3273551 (Why is no real title available?)
- Maximum likelihood estimation and large-sample inference for generalized linear and nonlinear regression models
- Natural exponential families with quadratic variance functions
- Natural real exponential families with cubic variance functions
- On polynomial variance functions
- Reproducibility and natural exponential families with power variance functions
- Stochastic approximation algorithms and applications
- The Lindsay transform of natural exponential families
- Semilinear credibility with several approximating functions
- Sequential credibility evaluation for symmetric location claim distributions
- Credibility theory: A new view from the theory of second order optimal statistics.
- Credibility theory and the Kalman filter
- A nonparametric sequential learning procedure for estimating the pure premium
- Sequential Quasi-Credibility for Scale Dispersion Models
- Contaminated Exponential Dispersion Loss Models
- Principal Applications of Bayesian Methods in Actuarial Science
- The structure of the distribution of a couple of observable random variables in credibility theory
This page was built for publication: On Stochastic Approximation and Credibility
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4258729)