Utility maximization and risk minimization in life and pension insurance (Q2488507)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Utility maximization and risk minimization in life and pension insurance
scientific article

    Statements

    Utility maximization and risk minimization in life and pension insurance (English)
    0 references
    0 references
    24 May 2006
    0 references
    The author considers an optimisation problem for a life insurance company or pension fund where the individual policy risk is taken explicitly into account. He works with a general participating or with a profit multi-state policy, characteristics of which are its contractual payment streams with built-in safety margins leading to the emergence of systematic surplus which then forms a source of bonus benefits paid to the policy holder in addition to the contractual benefits. The company faces an overall problem involving two concerns: optimisation on behalf of the insured and optimisation for the company itself according to suitable criteria. A two-step solution approach is proposed. The first step consists in utility maximization of the benefit stream for the insured through dividend allocation which is a widely used method of surplus redistribution. This step is based on a martingale methodology and convex analysis methods. The second step consists in treating the optimal benefit stream obtained at the first step as a given (random) liability on the part of the company and determining an invest strategy that takes this liability into account and is optimal for the company in a suitable sense. This leaves the company with a non-hedgeable liability, for which two quadratic hedging approaches are discussed. General results on dividend optimisation and semi-explicit expressions for the optimal bonus and investment strategies are obtained.
    0 references
    participating life insurance
    0 references
    dividends
    0 references
    bonus
    0 references
    investment strategies
    0 references
    optimisation
    0 references
    risk minimization
    0 references

    Identifiers