RUS  ENG
Full version
JOURNALS // Journal of Samara State Technical University, Ser. Physical and Mathematical Sciences // Archive

Vestn. Samar. Gos. Tekhn. Univ., Ser. Fiz.-Mat. Nauki [J. Samara State Tech. Univ., Ser. Phys. Math. Sci.], 2012 Issue 2(27), Pages 144–151 (Mi vsgtu1021)

Mathematical Modeling

Recurrent methods of construction of cumulative functions of risk

V. N. Nikishchov, E. V. Mikhailova

Samara State University, Samara, Russia

Abstract: The estimation problem of the enterprise total losses related to the improper performance of liability is considered. The total losses distribution function and density for the aggregate contracts are constructed by the recurrent methods, proposed by N. de Pril and H. Panjer. The numerical experiments based on five groups of contract portfolio are carried out. The results analysis showing the advantages and disadvantages of calculating algorithms is made. In particular, the values of means at risk obtained by H. Panjer method are superior to the corresponding values founded by N. de Pril method, and the maximum value of loss occurrence probability in the distribution density realized by H. Panjer method is less than the one, gotten by N. de Pril method. The results can be used for the further development of N. de Pril and H. Panjer approaches for estimation of total losses of risk aggregate.

Keywords: total loss, recurrent methods, cumulative functions of risk.

UDC: 519.864.3

MSC: Primary 91B30; Secondary 62P05

Original article submitted 24/XI/2011
revision submitted – 27/III/2012

DOI: 10.14498/vsgtu1021



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024