City Research Online

Credibility theory and experience rating in general insurance

Kandil, M.B.H. (1983). Credibility theory and experience rating in general insurance. (Unpublished Doctoral thesis, City University London)


The efficient use of available data to forecast future performance is one of the central concerns of modern business. In general insurance, especially, profit or loss depends upon charging the correct premium for the individual risk underwritten. When a new contract is signed, the insurer may know very little about the individual characteristics of the risk he has insured, and the premium charged is based upon prior statistics gathered from a large collective of similar, but also somewhat heterogeneous contracts grouped together under the same group rate. However, as the individual contract continues over several years, experience data can be gathered 'Co reflect individual risk characteristics. A simple but effective approach to estimating an experience rated premium is provided by credibility theory. Credibility theory
is concerned with the weight to be given to the small amount recent individual experience as compared with the mass of the data from the past collective experience. This research was undertaken to review different views and approaches to experience-rating in line with credibility theory considering fire insurance as an example of application ; also to develop a mathematical formula for experience rating for the pure premium keeping in mind- the conditions and the facilities available in the Egyptian market. The formula developed takes into account the variability of the claim amount and the claim frequency. Two classes of non-industrial fire insurance are analysed and tables for the credibility factors involved in the developed formula are constructed.

Publication Type: Thesis (Doctoral)
Subjects: Q Science > QA Mathematics
Departments: School of Science & Technology > Mathematics
PDF - Accepted Version
Download (6MB) | Preview



Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login