RISK PREDICTION OF CLAIM SEVERITY FRECHET AND GUMBEL DISTRIBUTION
Claim losses have become interesting studies in the insurance field because its related to the investment that big enough and have probabilistic, so that challenging to be studied deeper. There are two important aspects in the claim losses modelling, which are claim frequency and claim severity. In...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/20282 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Claim losses have become interesting studies in the insurance field because its related to the investment that big enough and have probabilistic, so that challenging to be studied deeper. There are two important aspects in the claim losses modelling, which are claim frequency and claim severity. In this thesis will be discussed about the claim severity data and model distribution. The appropriate candidates for <br />
<br />
<br />
<br />
modelling the claim severity are Frechet and Gumbel distributions. The risk measurement used to predict the amount of risk losses of claim severity are Value-at-Risk (VaR). The losses with extreme value happened in right tail distribution are very risky, because it will affect the company policies in determining the next insurance premium. Transformation to transfer the weight to the greater losses in the right tail distribution will be conducted using Proportional Hazard (PH) Transformation. Numerical simulation is conducted to predict the value of both measurement of VaR and VaR PH. From data analysis, the predicted value of VaR and VaR PH of Frechet distribution are greater than Gumbel distribution. |
|---|