@misc{Szkutnik_Włodzimierz_Składka_2005,
 author={Szkutnik, Włodzimierz},
 year={2005},
 rights={Wszystkie prawa zastrzeżone (Copyright)},
 publisher={Wydawnictwo Akademii Ekonomicznej im. Oskara Langego we Wrocławiu},
 description={Prace Naukowe Akademii Ekonomicznej we Wrocławiu; 2005; nr 1088, t. 2, s. 269-283},
 language={pol},
 abstract={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)},
 type={artykuł},
 title={Składka w aspekcie liniowej zależności korelacyjnej wielkości roszczeń i innych czynników},
}