Composite Rayleigh-Pareto distribution : Application to real fire insurance losses data set
Chaabane Benatmane, Halim Zeghdoudi, Rama Shanker, Nouara Lazri
Abstract
This paper proposes a new distribution, called as composite Rayleigh-Pareto (CRP) distribution. Various statistical properties like moment method and maximum likelihood estimation are established. Compared with the composite lognormal-Pareto distribution, the resulting density has a similar shape and a larger tail. This is why we expect that such a model will be a better fit than the composite lognormal-Pareto one for some heavy tailed insurance claims data. Simulated examples and fire insurance data sets (actual data on Algerian fire insurance losses, Danish fire loss data) were analyzed to show the importance and applicability of this new distribution.
Topics & Concepts
Pareto principleLog-normal distributionPareto distributionPareto interpolationLomax distributionRayleigh distributionStatisticsData setDistribution (mathematics)Generalized Pareto distributionMoment (physics)EconometricsComputer scienceMathematicsProbability density functionExtreme value theoryPhysicsClassical mechanicsMathematical analysisStatistical Distribution Estimation and ApplicationsInsurance, Mortality, Demography, Risk ManagementProbability and Risk Models