MODELING OF INSURANCE CLAIM COUNTS THROUGH JITTERS CONCEPT USING GAUSSIAN COPULA
Claim frequency is the number of claims within a period of time. Usefulness reviewing of claim counts is able to capture the characteristics of a risk. Here claim counts is reviewed that is longitudinal data. Intertemporal dependence of claims data can be captured by traditional models Random Effect...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/31166 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Claim frequency is the number of claims within a period of time. Usefulness reviewing of claim counts is able to capture the characteristics of a risk. Here claim counts is reviewed that is longitudinal data. Intertemporal dependence of claims data can be captured by traditional models Random Effects and Copula models. The model proposed in this research is modeling the intertemporal dependence of historical claims data through joint distribution using Copula. Copula used in this research is Gaussian Copula. Paramater of Gaussian Copula which represents intertemporal dependence is rho. The problem in this research is that claim counts are a discrete data. Meanwhile, the use of discrete data on Copula gives results that are not unique. To solve that problem it is used transformation from discrete data to continuous data. The research that is developed in this paper are modeling of joint distribution claim counts through jitters using Gaussian Copula, calculate predictions of claim counts in the future by reviewing the intertemporal dependence of claims data, and calculate premiums in the future by reviewing the intertemporal dependence of claims data. Correlation structure used in this research is the correlation structure AR1. Joint distribution of claim counts using Gaussian Copula applied to simulated data and real data. The results of this research are able to model the joint distribution of claim counts using Copula Gaussian, can calculate the predictions of claim counts in the future by reviewing the intertemporal dependence of claims data, and can calculate premiums in the future by reviewing the intertemporal dependence of claims data. |
|---|