ESTIMATION OF OUTSTANDING CLAIMS LIABILITY USING BORNHUETTER-FERGUSON METHOD AND SENSITIVITY ANALYSIS WITH LEVERAGE METHODOLOGY IN LONG-TAIL INSURANCE BUSINESS

• Main Author: Rahma Putri, Amelia
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/84143
• Institution: Institut Teknologi Bandung

A general insurance business consists of a long-tail and a short-tail insurance businesses. A long-tail insurance business is one in which the duration between the occurrence of an event which caused a claim to occur until the claims is finalized is more than a year. An incurred claim which has not been paid is called an Outstanding Claims Liability (OCL). Estimating an OCL is crucial as an insurance company needs to allocate sufficient funds to cover future claims. In this Final Project, the estimation of an OCL is carried out using a Bornhuetter-Ferguson (BF) method, which considers the pattern of past paid claims and the "premium" received. It is assumed that the premium equals to the estimated ultimate claims obtained using the Mack’s Chain Ladder (MCL) method. The sensitivity of the OCL estimate when there is an addition of an incremental claim in a particular cell of the runoff triangle, is analyzed using a measurement called the “Leverage”. The data used in this study is a runoff triangle of paid claims data from a general insurance company in Belgium. The results indicate that the BF method produces a higher OCL estimate compared to that produced by the MCL method. Moreover, the Mean Squared Error (MSE) of the leverage and the 0.995-quantile of the lognormal distribution of the resulting OCL are also determined.