Saddlepoint approximations to the distribution of the total claim amount in some recent risk models (Q4034592)

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 167704
Language Label Description Also known as
default for all languages
No label defined
    English
    Saddlepoint approximations to the distribution of the total claim amount in some recent risk models
    scientific article; zbMATH DE number 167704

      Statements

      Saddlepoint approximations to the distribution of the total claim amount in some recent risk models (English)
      0 references
      0 references
      16 May 1993
      0 references
      compound sum
      0 references
      conjugate distribution
      0 references
      log-concave density
      0 references
      tail probability
      0 references
      risk theory
      0 references
      total claim amount
      0 references
      saddlepoint approximation
      0 references
      Esscher approximation
      0 references
      claim process
      0 references
      Gaussian random variable
      0 references
      fluctuations in the premium
      0 references
      Poisson sums
      0 references
      inflationary conditions
      0 references
      Markov modulated risk process
      0 references
      claim amount distribution
      0 references
      lattice distribution
      0 references
      In the framework of risk theory the total claim amount is usually described by means of the formula: \(X(t)= \sum^{N(t)}_{i=1} X_ i\), where \(N(t)\) represents the number of claims, \(N(t)\sim\text{Poisson} (\beta t)\), and the claims \(X_ i\) are independent and identically distributed and also independent of \(N(t)\). A well-known approximation for the total claim distribution is the saddlepoint approximation or Esscher approximation. The purpose of the paper is the extension of the above mentioned approximations to other risk models. At first, the author extends the saddlepoint approximation for \(P(X(t)>x)\) in the case of a claim process of the form: \[ X(t)= U(t)+ \sum^ d_{k=1} \sum^{N(t;k)}_{i=1} X_{k_ i}\quad\text{(model of K. K. Aase)}, \] where \(U(t)\) is a Gaussian random variable related to fluctuations in the premium, and the last term on the right hand side is a sum of \(d\) ordinary Poisson sums. Then, using \textit{G. E. Willmot's} results in ibid. 1989, No. 1, 1-12 (1989; Zbl 0679.62094), he investigates the classical model under inflationary conditions, and finally the Markov modulated risk process, which was proposed by \textit{S. Asmussen} [see ibid. 1989, No. 2, 69-100 (1989; Zbl 0684.62073)] is studied. The results concerning the mentioned models are stated by assuming that the claim amount distribution has a density with respect to the Lebesgue measure; in the final part of the paper the case where the claim distribution is a lattice distribution is discussed.
      0 references

      Identifiers