×
Home Current Archive Editorial board
News Contact
Scientific article

MODELIRANJE ŠTETA U OSIGURANJU PRIMENOM MEŠAVINE LINEARNIH EKSPONENCIJALNIH RASPODELA

By
Vesna Rajić ,
Vesna Rajić

Univerzitet u Beogradu, Ekonomski fakultet Beograd , Beograd , Serbia

Tatjana Rakonjac-Antić
Tatjana Rakonjac-Antić

Univerzitet u Beogradu, Ekonomski fakultet Beograd , Beograd , Serbia

Abstract

Određivanje raspodela verovatnoća šteta je veoma značajno za osiguravajuće kompanije, jer se na osnovu njih određuju premije osiguranja, rezerve, solventnosni kapital i samopridržaj kompanije. U radu se predlaže primena mešavine linearnih eksponencijalnih raspodela za modeliranje šteta u osiguranju. Ocene po metodu maksimalne verodostojnosti parametara ove raspodele se dobijaju korišćenjem EM algoritma. Izvedeni su obrasci za izračunavanje VaR i TVaR.

References

1.
Bilmes J. A Gentle Tutorial of the EM Algorithm and its Application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models. 1998;
2.
Bohning D. 1999;
3.
Dempster A, Laird N, Rubin D. Maximum-likelihood from incomplete date via the em algorith. J Royal Statist Soc Ser B. 1977;1–38.
4.
Hardy M. An Introduction to Risk Measures for Actuarial Applications. CAS Exam Study. 2006;1–31.
5.
Kim SH. Calibrated initials for an EM applied to recursive models of cstegorical variable. Computational Statistics & Data Analysis. 2002;97–110.
6.
Kočović J, Rajić V, Jovanović M. Estimating a Tail of the Mixture of Lognormal and Inverse Gaussian Distribution. Scandinavian Actuarial Journal. 2013;10.
7.
Kočović J, Šulejić P, Rakonjac-Antić T, Osiguranje. Centar za izdvačku delatnost, Ekonomski fakultet. 2010;
8.
Linsday B. Mixture Models: Theory, Geometry and Application. 1995;
9.
Mclachlan G, Peel D. 2000;
10.
Rajić V, Stanojević J. Mešavina linearnih eksponencijalnih raspodela. 2010;739–42.
11.
Woodward W, Parr W, W, Schucany R, Lindsey H. A comparision of minimum distance and maximum likelihood estimation of mixture proportio. J Amer Statist Assoc. 1984;590–8.

Citation

Article metrics

Google scholar: See link

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.