A generalized linear model for rating the premiums of a Micro Health Insurance Policy

Abstract

This study focuses on the modeling of a risk premium for a micro health insurance policy. It was carried out as a response to the gap highlighted by scholars and industry players which relates to the difficulty in developing insurance products which are both affordable and sustainable to cater for the needs of the low income market. which is a large segment in developing countries (KPMG, 20 13). It adopts the generalized linear model in the rating of premium factors to parameterize their Significance in predicting the claim amount. The product being analyzed is a group micro health insurance policy in the Kenyan market, and claims data has been .obtained from three .schemes whose policies have been in force for at least three years since 2011. A policy covers not only the scheme's members but also their nuclear family. Factors rated include the claimant's age, gender, and their relation to the insured life. The impact of the interaction between these factors is also analyzed. The results show that for a group policy offering family cover, all the three factors are significant in the prediction of claim amounts and thus the risk premium. They also show that interactions between the claimant's age and gender and between their age and relation to the policy holder significantly improve the pricing model.