Journal of Mathematical Sciences

, Volume 93, Issue 4, pp 552–562 | Cite as

Comparison of ruin probability estimates in the presence of heavy tails

  • D. G. Konstantinidis
Article

Abstract

Four methods of estimation of the ruin probability in the presence of heavy tails are compared in accordance with their effectiveness on the Pareto-distributed claim sizes. The first method, proposed by Embrechis and Veraverbeke, provides an asymptotic expression when the initial capital tends to infinity. The second method, proposed by Goovaerts and De Vylder, provides an algorithm for two-sided estimation based on the solution of the renewal equation through discretization. Its advantage is the perfect applicbility in cases with small initial capital. The third method, proposed initially by Willmoi and improved by Kalashnikov, provides an upper bound of the ruin probability with the help of a test function. The last method, proposed by Kalashnikov, provides two-sided bounds for the ruin probability using truncation of random variables to avoid the Cramér condition. The last two methods enable one to handle cases with intermediate initial capital.

Keywords

Probability Estimate Asymptotic Expression Heavy Tail Initial Capital Claim Size 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    F. Dufresne and H. U. Gerber, “Three methods to calculate the probability of ruin,”AUSTIN Bull.,19, 71–90 (1989).Google Scholar
  2. 2.
    P. Embrechts, C. M. Galled, and N. Veraverbeke, “Subexponentiality and infinite divisibility,”Z. War. verb. Gibbet,49, 335–347 (1979).Google Scholar
  3. 3.
    P. Embrechts and N. Veraverbeke, “Estimates for the probability of ruin with special emphasis on the probability of large claims,”Insurance: Math. and Econ.,1, 55–72 (1982).MathSciNetGoogle Scholar
  4. 4.
    M. J. Goovaerts and F. De Vylder, “A stable algorithm for calculation of ultimate ruin probability,”AUSTIN Bull.,14, 53–59 (1984).Google Scholar
  5. 5.
    J. Grandell,Aspects of Risk Theory, Springer, New York (1991).Google Scholar
  6. 6.
    V. Kalashnikov,Topics on Regenerative Processes, CRC Press, Boca Raton (1994).Google Scholar
  7. 7.
    V. Kalashnikov, “Bounds for geometric sums in the presence of heavy-tailed summands,” in:Poépublications de Lockup Denials et de Matématiques Appliquées,27, Université de Maine-La-Valley (1995).Google Scholar
  8. 8.
    V. Kalashnikov, “Two-sided bounds of ruin probabilities,”Scand. Act. J.,1, 1–18 (1996).MATHMathSciNetGoogle Scholar
  9. 9.
    V. Kalashnikov,Geometric Sums: Bounds for Rare Events with Applications, Clear Acid. Publ. (1997) (in preparation).Google Scholar
  10. 10.
    V. Kalashnikov and D. Konstantinidis, “Ruin probability,”Fund. Prikl. Mat.,2, No. 4 (1996).Google Scholar
  11. 11.
    H.-J. Rossberg and G. Siegel, “Die bedeuting von kingmans integralunggleichungen bei der approximation der stationären wartezeitverteilung im modellGI |G| 1 mit und ohne verzögerung beim beginn einer beschäftigungsperiode,”Math. Oper. Statist.,5, 687–699 (1974).MathSciNetGoogle Scholar
  12. 12.
    G. Willmot, “Refinements and distributional generalizations of Lundberg's inequality,”Insurance: Math. Econ.,15, 49–63 (1994).MATHMathSciNetGoogle Scholar

Copyright information

© Kluwer Academic/Plenum Publishers 1999

Authors and Affiliations

  • D. G. Konstantinidis
    • 1
  1. 1.Department of MathematicsTechnical University of CreteChaniaGreece

Personalised recommendations