DETERMINING RESERVE OF THRESHOLD IN A RANDOM VARIABEL FOR EXTREM CLAIM SEVERITY

• Main Author: Suciana Arnel, Silfia
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/50193
• Institution: Institut Teknologi Bandung

An insurance company need to manage the insurance risk to avoid the error in measuring the risk. The substantial risk faced by insurance policyholders is referred to as a claims severity. The significant number of claims with extreme large claims may occur. In this thesis, claims severity data is analyzed using an Extreme Value Theory (EVT) model with the Peak Over Threshold (POT) approach. In specifying an extreme value using the POT method will assigned some candidates of a limit of threshold. Threshold obtained from fixed variable is specified with mean excess function plot. While the Threshold obtained from random variables is determined using the minimum upper limit of the claim severity for each year.The threshold candidate will be tested using the Kolmogorov Smirnov test to get a threshold value that matches with the Generalized Pareto distribution. The matching value will be used in performing calculating premium and a reserved claim. In applying this method, will use broad data claims on property insurance that occurred in the year 2010-2016.