Scholarly record

MODELLING OF EXTREME DAMAGES IN INSURANCE

Bohdan Linda, V. Pacakova, Z. Zackova

Abstract

Insurance means protection against different kinds of risks with financial consequences. In frame of this protection, the insured transfers its risks, the possible negative consequences of which are unbearable to the insurer, i.e. to an insurance company. In order to compensate the insurer for such consequences, the insurance company collects the premium for the transfer of risk, the amount of which depends on the size of conventional damages, having a random character. The balance between premium written and premium settled may disrupt extreme damages, which are rare, but they represent a huge financial burden for the insurance company. Due to the infrequent occurrence of these events it is difficult to estimate their probability distributions by exact statistical methods, requiring the fulfillment of various assumptions. In practice, probability distributions, such as Weibull's distribution, Pareto's distribution, Lognormal distribution, Log-logistic distribution, Generalized extreme value distribution and others with heavy right tail, are often used. Simulation techniques allow modeling their distribution without the need to verify assumptions. The paper deals with the application of Monte Carlo simulation methods for determining the size of the potential catastrophic damage, which may be used by insurers to determine the limit of claims, which are transferred to reinsurance.

Publication Impact Profile

No metrics available.

PLUMX see details

DOI resolver: doi.org/10.5593/sgemsocialF2018/1.6/S03.060

Publication details

Back to publication list