Modeling occurrence of death in Ghanaian female pensioners

Thumbnail Image
October 2016
Journal Title
Journal ISSN
Volume Title
The study considered count data with excess zeros and over-dispersion. The models used for this situation of the study data are the zero-inflated negative binomial (ZINB) and the negative binomial (NB) regression model. This was to establish whether the parameter age is a function of mortality for the various cohort groups. Also the Akaike Information Criteria (AIC) goodness of fit testing was used to determine a best fit for the cohort groups with the various model. The ZINB regression model best fitted the female pensioners‟ data for the 2005 cohort group while the NB regression model best fitted the 1990 cohort group. Other parameters like causes of death and standard of living after retirement must be considered in future model.
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 Master of Science in Actuarial Science,