This item still needs to be validated !
Title: Combining Panjer's Recursion with Convolution
Authors: Kaas, Robert ×
Van Heerwaarden, AE
Goovaerts, Marc #
Issue Date: 1989
Series Title: Insurance: Mathematics & Economics vol:8 issue:1 pages:19-21
Abstract: Distribution function and stop-loss premiums of total claims S on an insurance portfolio can be computed using convolution, which is a slow but exact method. The number of computations needed is proportional to the total number of mass points of all terms multiplied by the maximum of the range in which the values of the distribution function and the stop-loss premiums are needed. Panjer's recursion is a fast algorithm for a model having a compound Poisson distribution, but the probabilities obtained are not exactly equal to those of S. A method is proposed that combines the good points of both methods. Convolution is used only for the important parts of the risk, with less important parts replaced by the Poisson approximation. The proposed model for the total claims on a nonlife portfolio is almost as attractive from a computational point of view as the compound Poisson model. Its stop-loss premiums are much better upper bounds for the exact stop-loss premiums
ISSN: 0167-6687
Publication status: published
KU Leuven publication type: IT
Appears in Collections:Research Center Insurance, Leuven
× corresponding author
# (joint) last author

Files in This Item:

There are no files associated with this item.

Request a copy


All items in Lirias are protected by copyright, with all rights reserved.