Czasopismo
2005
|
nr 1088, t. 2 Inwestycje finansowe i ubezpieczenia - tendencje światowe a polski rynek. T. 2
|
269--283
Tytuł artykułu
Autorzy
Warianty tytułu
Premium in the Aspect of Linear Dependence Correlative Claims' Size and Other Factor
Języki publikacji
Abstrakty
Przedstawiono sposób wyznaczania regresji składkowej względem wielkości roszczeń oraz omówiono związek między rozstępem w poziomie roszczeń a rozstępem w poziomie składek.
The theme in the article is to show, in its main dimension, that various approaches may by applied to the analysis of the issue concerning relation between estimated premium and distribution of insurance claims'probability.
The models presented in the article, in spite of some simplified scheme of its exemplification in the linear form, are far-reaching departure from the assumptions assumed, for instance, in econometric modeling.
The standpoint proposed in the article allows to define the estimation of premium P in the form of sum (total) of claims'size, assuming their invariability, measured with premium regression coefficient in relation to claims and other factors affecting the claim risk and the second ingredient which may be interpreted as an additional claim.
The far-reaching generalization was made as well, assuming that the quantity intensifying the influence of random claims on the random claim is random as well, it means that the constant from the models considered earlier is now a random variable.
The purpose of the considerations is also achievement of both estimation of variance of the ingredient additionally affecting (apart from claims) premium and of variance of this "randomnised" constant. The notion of the so-called variances of conditional averages. It is shown that variance of negotiated compensation, what implies a certain a-tipicality of the considered insurance procedures, will be in all cases not bigger then the sum of average error at the old information and the error expressed by the variance of conditional averages the new information. (original abstract)
The models presented in the article, in spite of some simplified scheme of its exemplification in the linear form, are far-reaching departure from the assumptions assumed, for instance, in econometric modeling.
The standpoint proposed in the article allows to define the estimation of premium P in the form of sum (total) of claims'size, assuming their invariability, measured with premium regression coefficient in relation to claims and other factors affecting the claim risk and the second ingredient which may be interpreted as an additional claim.
The far-reaching generalization was made as well, assuming that the quantity intensifying the influence of random claims on the random claim is random as well, it means that the constant from the models considered earlier is now a random variable.
The purpose of the considerations is also achievement of both estimation of variance of the ingredient additionally affecting (apart from claims) premium and of variance of this "randomnised" constant. The notion of the so-called variances of conditional averages. It is shown that variance of negotiated compensation, what implies a certain a-tipicality of the considered insurance procedures, will be in all cases not bigger then the sum of average error at the old information and the error expressed by the variance of conditional averages the new information. (original abstract)
Rocznik
Strony
269--283
Opis fizyczny
Twórcy
autor
- Akademia Ekonomiczna im. Karola Adamieckiego w Katowicach
Bibliografia
- David A., Controling Premia by Repackaging Asset-Backet Securities, "The Journal of Risk Insurance" 1997, vol. 64, nr 4, s. 619-648.
- Downs D.H., Monitoring, Ownership, and Risk-Taking: the Impact of Guaranty Funds, "The Journal of Risk Insurance" 1997, vol. 66, nr 3, s. 477-497.
- Feller W., Wstęp do rachunku prawdopodobieństwa, PWN, Warszawa 1966.
- Fisz M., Rachunek prawdopodobieństwa i statystyka matematyczna, PWN, Warszawa 1969.
Typ dokumentu
Bibliografia
Identyfikatory
Identyfikator YADDA
bwmeta1.element.ekon-element-000171546579
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.