One of the most frequently recurring subjects found in actuarial literature is the problem of how to assess the risk associated with a portfolio of life policies. Almost all the studies are based on mathematical treatments which, by making various assumptions, provide solutions to the problem mentioned. A number of contributions published recently have suggested that the assessment of risk can be based on the construction of general algorithms; in these stidues, estimates are made - in particular - for the value of ruin, the year of its occurrence and the probability distribution of the capital prior to it. In other words, by contrast to the usual mathematical approch which, by making various assumptions , attempts to carry out valuations that are independent on the specific case, this paper proposes a different approach based on an algorithm that is founded on general hypoteses: an algorithm that requires just the single risk variable for the portfolio concerned to be defined. The last part of this study develops a treatment of dependent risk variables that avoinds calculating the covariances.
The direct and exact assessment of actuarial risk: application with regard to portfolio of policies / Attias, Anna; S., Tumani. - STAMPA. - (2000), pp. 3-14. (Intervento presentato al convegno XXXI Internation Astin colloquium tenutosi a Porto Cervo nel 17-20 settembre 2000).
The direct and exact assessment of actuarial risk: application with regard to portfolio of policies
ATTIAS, Anna;
2000
Abstract
One of the most frequently recurring subjects found in actuarial literature is the problem of how to assess the risk associated with a portfolio of life policies. Almost all the studies are based on mathematical treatments which, by making various assumptions, provide solutions to the problem mentioned. A number of contributions published recently have suggested that the assessment of risk can be based on the construction of general algorithms; in these stidues, estimates are made - in particular - for the value of ruin, the year of its occurrence and the probability distribution of the capital prior to it. In other words, by contrast to the usual mathematical approch which, by making various assumptions , attempts to carry out valuations that are independent on the specific case, this paper proposes a different approach based on an algorithm that is founded on general hypoteses: an algorithm that requires just the single risk variable for the portfolio concerned to be defined. The last part of this study develops a treatment of dependent risk variables that avoinds calculating the covariances.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.