COMPOUND POISSON MODEL: SPECIAL CASE OF TWEEDIE MODEL
Claims experience in a long-tail insurance business may be recorded on a Run-O Triangle (ROT). One of the methods that can be used to predict claim in ROT is chain ladder method, which is based on some assumptions. The thesis describes the dierences between chain ladder recursive model and chain...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/34005 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Claims experience in a long-tail insurance business may be recorded on a Run-O
Triangle (ROT). One of the methods that can be used to predict claim in ROT is
chain ladder method, which is based on some assumptions. The thesis describes
the dierences between chain ladder recursive model and chain ladder non-recursive
model with each own parameter estimation.
The thesis also consider Tweedie distribution family, especially compound Poisson
with Gamma severity as the distribution of an incremental claim in a cell of
a particular ROT which leads to Tweedie's compound Poisson model. One of the
members of Tweedie distribution family, that is over-dispersed Poisson distribution,
is used as a probability model underlying the chain ladder, recursive and nonrecursive,
models. It is also demonstrated that the maximum likelihood estimation
for the parameters of Tweedie's compound Poisson model can be obtained using the
chain ladder non-recursive model algorithm for parameter estimation. |
|---|