Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter
When modelling positively skewed insurance claim data, traditional distributions such as lognormal and Weibull often fail to accurately estimate the tail. Several methods have been developed to improve tail estimation without compromising the body fitting, including the transformed kernel density an...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/44131 |
| _version_ | 1811690258592432128 |
|---|---|
| author | Seet, Angeline Yuen Chee Yang, Bowen Yeoh, Yun Wei |
| author2 | Uditha Balasooriya |
| author_facet | Uditha Balasooriya Seet, Angeline Yuen Chee Yang, Bowen Yeoh, Yun Wei |
| author_sort | Seet, Angeline Yuen Chee |
| collection | NTU |
| description | When modelling positively skewed insurance claim data, traditional distributions such as lognormal and Weibull often fail to accurately estimate the tail. Several methods have been developed to improve tail estimation without compromising the body fitting, including the transformed kernel density and the generalised lambda distribution. In this study, we investigate the robustness of a promising method, fitting of the hyper-Erlang distribution with a common scale parameter. A modified version of the Expectation Maximisation (EM) algorithm is used for distribution fitting, with some changes we proposed to improve the efficiency of the algorithm. Results from a preliminary study we conducted suggest that different initial estimates of the common scale parameter affect the performance of the modified EM algorithm. For fitting medical claim data provided by the Society of Actuaries, bootstrap samples are taken to determine an optimal initial estimate for the scale parameter. With this estimate, the hyper-Erlang distribution is able to provide a satisfactory fit to the data. The result is comparable to those produced by transformed kernel density and generalised lambda distribution. |
| first_indexed | 2024-10-01T06:01:08Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/44131 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:01:08Z |
| publishDate | 2011 |
| record_format | dspace |
| spelling | ntu-10356/441312023-05-19T06:16:16Z Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter Seet, Angeline Yuen Chee Yang, Bowen Yeoh, Yun Wei Uditha Balasooriya Nanyang Business School DRNTU::Business::Finance::Insurance claims When modelling positively skewed insurance claim data, traditional distributions such as lognormal and Weibull often fail to accurately estimate the tail. Several methods have been developed to improve tail estimation without compromising the body fitting, including the transformed kernel density and the generalised lambda distribution. In this study, we investigate the robustness of a promising method, fitting of the hyper-Erlang distribution with a common scale parameter. A modified version of the Expectation Maximisation (EM) algorithm is used for distribution fitting, with some changes we proposed to improve the efficiency of the algorithm. Results from a preliminary study we conducted suggest that different initial estimates of the common scale parameter affect the performance of the modified EM algorithm. For fitting medical claim data provided by the Society of Actuaries, bootstrap samples are taken to determine an optimal initial estimate for the scale parameter. With this estimate, the hyper-Erlang distribution is able to provide a satisfactory fit to the data. The result is comparable to those produced by transformed kernel density and generalised lambda distribution. BUSINESS 2011-05-26T07:31:08Z 2011-05-26T07:31:08Z 2011 2011 Final Year Project (FYP) http://hdl.handle.net/10356/44131 en Nanyang Technological University 85 p. application/pdf |
| spellingShingle | DRNTU::Business::Finance::Insurance claims Seet, Angeline Yuen Chee Yang, Bowen Yeoh, Yun Wei Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title | Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title_full | Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title_fullStr | Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title_full_unstemmed | Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title_short | Modelling heavy-tailed insurance claim data using the hyper-erlang distribution with common scale parameter |
| title_sort | modelling heavy tailed insurance claim data using the hyper erlang distribution with common scale parameter |
| topic | DRNTU::Business::Finance::Insurance claims |
| url | http://hdl.handle.net/10356/44131 |
| work_keys_str_mv | AT seetangelineyuenchee modellingheavytailedinsuranceclaimdatausingthehypererlangdistributionwithcommonscaleparameter AT yangbowen modellingheavytailedinsuranceclaimdatausingthehypererlangdistributionwithcommonscaleparameter AT yeohyunwei modellingheavytailedinsuranceclaimdatausingthehypererlangdistributionwithcommonscaleparameter |