Analysis on risk factors affecting the likelihood and severity of motor insurance claims in Singapore

Abstract

This project investigates the risks factors used in the pricing of motor insurance in Singapore. We study the impact of various risk factors on the likelihood of incurring a claim and the claim severity, namely Age of Insured Vehicle, Engine Capacity of Insured Vehicle, No-Claim-Discount Status of Policyholder, Gender of Policyholder and Age of Policyholder. The analysis is done using 1,934 in-force private cars insurance policy claims from 1998 to 2000 obtained from NTUC Income, the largest motor insurer in Singapore. Binary Logistic Regression is used to model the probability of incurring a claim due to the dichotomous nature of our dependent variable and General Linear Model (GLM) is used to model the amount of claim incurred, a scale variable. Our results conclude that Age of Insured Vehicle, Engine Capacity of Insured Vehicle and No-Claim-Discount Status of Policyholder have significant impact on the probability of a incurring a claim. However, only the Age of Insured Vehicle and No-Claim-Discount Status of Policyholder have an impact on the amount of claim incurred. Since these risk factors could potentially affect the claims frequency and severity, a greater emphasis should be placed on such factors when considering the pricing issue of motor insurance policies.