A suitable claim severity model of comprehensive insurance policy
No Thumbnail Available
Date
2016-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
KNUST
Abstract
This study was base on a claim data collected from a recognized motor insurance
companies in Ghana. The data from this non-life insurance company was to find the
appropriate statistical distributions that best fit the claims data. In the insurance
industry two major problems exist; how to get an appropriate model for a large claim
data and how to fit and test the appropriateness of the model identified to make
future decision. In this study, a clear methodology was set up to deal with the
identified problems. Specific distributions were chosen from a group of family of
exponential distribution. The study began with a brief introduction for the modeling
process as practice by Actuaries. The purpose of the study was to get the best
statistical distribution that best fit the claim amount from the insurance company
data and fit the data using Quantile–Quantile plot (Q-Q plots) for all the tested
distributions. The modeling process established one of the statistical distributions
efficiently fitted the claim data. The study went ahead to test for the goodness of fit
test using (AIC) that is Akaike’s Information Criterion and a graphical presentation
was done using the Q-Q plot. Finally the findings were discussed and conclusions, was
made to the work and recommendations were given to guide for further study on
similar or same topic.
Description
A thesis submitted to the Department of Mathematics,
Kwame Nkrumah University of Science and Technology in
Partial fulfillment of the requirement for the Degree of MSc. Actuarial Science