Description: We find the necessary and sufficient conditions for fitting five given moments by a triatomic distribution. We consider three examples drawn from fire large spread, individual life medium spread and group life small spread insurance data, fit them with triatomics, and compute the ruin probabilities using well known formulas for discrete and for combination of exponentials claim amounts. We then compare our approximations with the exact values that appeared in the literature.

Description: In this paper, given the three moments of the claim amount, the authors find the di-atomic and the di-exponential distributions that fit the given three moments. They use ruin probability formulas when the claim amount is discete or is a combination of exponentials to compute ruin probabilities based on the claim amount distribution which match the first three moments of the original claim amount distribution. Actuarial Research Clearing House 1992 VOL. 1

Description: This paper considers several features in a model that experiences gradual drifts and jumps for prevalence and survivorship as is the case for AIDS. It is an actuarial model for AIDS when transition intensities follow jump-diffusion processes.

Description: In this note, we rearrange the statement and the proof of a theorem in Greville. The presentation appears to be more compact the materials, however, are not new.].

Description: To solve multiple choice problems by elimination, one approach is to study values of functions at the boundary. In this note, the author shows by examples that the method of boundary values can be risky if the tacit assumption of continuity at the boundary is not checked. From ACTUARIAL RESEARCH CLEARING HOUSE, 1980 VOL. 2.

Description: This paper compares percentile estimates of a large number of cash flow testing scenarios, where only a very small number fall in the ruin tail, by parametric and non-parametric methods.