Prediction of mortality rates using augmented data
Prediction of future mortality rate is of significant priority in the insurance industry today as insurers face challenging tasks in providing retirement benefits to a population with increasing life expectancy. A time series model based on multivariate power-normal distribution has been used in the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Teknologi Malaysia Press
2016
|
| Subjects: | |
| Online Access: | http://eprints.sunway.edu.my/432/1/Pooi%20Ah%20Hin%202.pdf http://eprints.sunway.edu.my/432/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Prediction of future mortality rate is of significant priority in the insurance industry today as insurers face challenging tasks in providing retirement benefits to a population with increasing life expectancy. A time series model based on multivariate power-normal distribution has been used in the literature on the United States (US) mortality data in the years 1933 to 2000 to predict the future mortality rates in the years 2001 to 2010. To improve the predictive ability, the US mortality data is augmented to include more variables such as death rates by gender and death rates of other countries with similar demographics. Apart from having good ability to cover the observed future mortality rate, the prediction intervals based on the augmented data performed better because they also tend to have shorter interval lengths. |
|---|