Churn Model in the Mobile Telecommunications Industry in Ghana: A Case Study of Vodafone Ghana

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
In this study, the purpose was to propose an approach to assist Telecom Churn Management, use Kaplan Meier estimator to estimate survival probabilities for each of the value bands High Value Customer (HVC), Medium Value Customer (MVC) and Low Value Customer (LVC), to test whether there is differences in the survival curves of the three value bands and to model the effect of usage (Total MOU, GPRS Usage, Total SMS and Top up Amount) on the risk of churning using Cox Proportional Hazard Model. Data on Prepaid customers was obtained from the start of SIM activation to date. The Kaplan Meier estimator and Cox Proportional Hazard model were the Statistical methods used in the analysis. From the result, we observed that the median churn time for High Value Customers (HVC) is approximately 20 months, Medium Value Customers have a median churn time of 25 months and the Low Value Customers have 11 months of churn median time. Generally, the Log-Rank test indicated significant differences among the survival function of the three value bands (HVC, MVC and LVC). From the Cox model total minute of use (Total MOU) and Top up Amount have significant effects on the survivorship. We conclude that customers who have high top up amount have low risk of churning and have longer survival times.  
A thesis submitted to the Institute of Distance Learning, Kwame Nkrumah University of Science and Technology in partial fulfilment of the requirement for the award of Master of Science degree in Industrial Mathematics,October-2013