Modeling lapse rates with economic variables: the vector autoregressive model approach
No Thumbnail Available
Date
2016-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
KNUST
Abstract
Life insurance contracts are accompanied by risks. This study focuses on the risk
involving termination of life policies by the policyholders. This study seeks to
identify the main determinants of lapse rates in the Ghanaian life insurance
industry. The data available span from 2006 to 2014 and were recorded on
monthly basis. Literature research on predictive modeling of lapse rates in the
insurance industry led to the choice for predicting with a VAR Model. A total of
five (5) variables were analyzed on their relationship with the average lapse rate.
Of these variables inflation, stock market return and interest rate proved
valuable for modeling. The vector auto-regression (VAR) model was used to
capture the evolution and the interdependencies between the variables. All the
variables in the VAR were treated symmetrically by including for each variable
an equation explaining its evolution based on its own lags and the lags of all the
other variables in the model. The main limitation of this research was the small
amount of data and their level of detail. Since the results of this research are
promising it is recommended to extend the research to other parts of the country.
To increase the statistical strength and accuracy of the inferences that can be
made it is also recommended to examine the lapse rates at a policyholder level.
Description
A thesis submitted to the Department of Mathematics,
Kwame Nkrumah University of Science and Technology in partial
fulfillment of the requirements for the award of MSc. Actuarial Science.