On Stochastic Approximation and Credibility (Q4258729)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1336594
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | On Stochastic Approximation and Credibility |
scientific article; zbMATH DE number 1336594 |
Statements
On Stochastic Approximation and Credibility (English)
0 references
14 September 1999
0 references
fair premium
0 references
credibility formulas
0 references
stochastic approximation
0 references
exponential dispersion
0 references
optimal stepwise gain sequence
0 references
location dispersion family
0 references
0 references
0.840214729309082
0 references
0.8199769854545593
0 references
0.8175949454307556
0 references
0.792273759841919
0 references
0.78731769323349
0 references
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.
0 references