MODELING OF VEHICLE INSURANCE PREMIUMS BASED ON CLAIM HISTORY WITH USING POISSON DISTRIBUTION
The success of a vehicle insurance company can be seen by the right determination of pure premium. In this research, the determination of pure premium by modeling claims history collection in the form of panel data. Modeling <br /> <br /> process of claims history data consider the chara...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/29702 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | The success of a vehicle insurance company can be seen by the right determination of pure premium. In this research, the determination of pure premium by modeling claims history collection in the form of panel data. Modeling <br />
<br />
process of claims history data consider the characteristics of the policyholder during the period. The data used in this research is vehicle insurance data, <br />
<br />
especially non-bus and non-truck data who claimed in 2011 and 2012. Based on the data obtained some characteristics that can be observed, such as sex, type of car, and the year of car manufactured is assumed that could affect the <br />
<br />
occurrence of claims. While the characteristics that cannot be observed, is called the random effects such as driving habits, consume drugs and alcohol. In this research, to calculate the claim frequency and pure premium using the Poisson distribution given Gamma distributed random effect thus obtained Poisson-Gamma models the claim frequency and pure premium for the next <br />
<br />
period. The value of the parameter Poisson-Gamma models are estimated using Maximum Likelihood Estimation. Furthermore, the estimates of the parameters used in the calculation of pure premiums coming period. |
|---|