Search

To model statistical data coming from two di erent distributions, Cooray and Ananda [1] introduced a composite (two-spliced) Lognormal-Pareto model, that was further extended by Scollnik [9] and fitted to insurance data. Inspired by these studies, more general three-spliced composite models are considered in this work, built by joining three di erent distributions. In particular, the study is focused on the three-spliced Exponential-Lognormal-Pareto distribution. The main characteristics of this model, as well as statistical inference are discussed. The parameters estimation is illustrated on random generated data.

The Fast Fourier Transform provides an alternative approximate method to evaluate the distribution of aggregate losses in insurance and finance. The efficiency of this method has already been proved for univariate and bivariate insurance models; therefore, in this paper, we extend it to a multivariate setting by considering its application to a particular model that includes losses of different types and dependency between them. Since the Fourier transform method works with truncated claims distributions, it can generate aliasing errors by wrapping around the probability mass that lies at the truncation point below this point. To eliminate this problem, we also discuss a suitable change of measure called exponential tilting that forces the tail of the distribution to decrease at exponential rate. Other possible errors are also discussed. We also illustrate the method on several numerical examples.

The discretization of a continuous distribution is an important and challenging step in problems emerging from different fields, like, e.g., finance and insurance. The usual discretization methods come along with some information loss. With the purpose to capture more information by covering the entire support of the original distribution while keeping the discrete characteristic, in this paper, we propose a fuzzy alternative discretization method that—in terms of random variables—replaces a discretized random variable with a discrete triangular fuzzy random variable. As applications, we insert this discrete triangular fuzzy random variable in two classical risk models and study the results.

In this work, we consider the multivariate aggregate model introduced in [11], model that takes into account the case when different types of claims affect in the same time an insurance portfolio under some specific assumptions related to the number of claims. For the probability function of the corresponding multivariate compound distribution, [11] obtained an exact recursive formula proved using the properties of the probability generating function. In this paper, we present a new shorter proof of the same formula that we also extend to a new form. Moreover, we present an alternative approximate method to evaluate the compound distribution based on the Fourier transform, and we compare both methods on a numerical example.

Recently, nonhomogeneous claim sizes have been considered in the actuarial literature starting from the fact that the claims are seasonally influenced by the economic environment. In this context, Raducan et al. [8] obtained recursive formulas for the ruin probability at or before claim instants, and stated a conjecture that relates the order of the claims arrival to the magnitude of the corresponding ruin probability, conjecture supported by numerical examples. In this paper, we prove this conjecture in a particular case.

Abstract: We consider a discrete-time insurance risk model, in which the financial risks constitute a stationary process with finite dimensional distributions of Farlie–Gumbel–Morgenstern type. We obtain an exact asymptotic formula for the ruin probability, reflecting the impact of this kind of association structure among the financial risks. [Copyright &y& Elsevier]

Abstract: In the actuarial literature, several exact and approximative recursive methods have been proposed for calculating the distribution of a sum of mutually independent compound Bernoulli distributed random variables. In this paper, we give an overview of these methods. We compare their performance with the straightforward convolution technique by counting the number of dot operations involved in each method. It turns out that in many practicle situations, the recursive methods outperform the convolution method.

AbstractIn the present paper, we study error bounds for approximations to multivariate distributions. In particular, we discuss some general versions of compound multivariate distributions and look at distributions of dependent random variables constructed by linear transforms of independent random variables or vectors. Special attention is paid to the case when the support of the original distribution is restricted. We also look at some applications with multivariate Bernoulli distributions.

AbstractThis paper proposes a multivariate generalization of the generalized Poisson distribution. Its definition and main properties are given. The parameters are estimated by the method of moments.

AbstractIn this paper we consider compound distributions where the counting distribution is a bivariate distribution with the probability function (Pn1,n2)n1,n2=0that satisfies a recursion in the formWe present an algorithm for recursive evaluation of the corresponding compound distributions and some examples of distributions in this class.

AbstractThis paper deals with the bivariate generalized Poisson distribution. The distribution is fitted to the aggregate amount of claims for a compound class of policies submitted to claims of two kinds whose yearly frequencies are a priori dependent. A comparative study with the bivariate Poisson distribution and with two bivariate mixed Poisson distributions has been carried out, based on data concerning natural events insurance in the USA and third party liability automobile insurance in France.