Mortality Modelling And Forecasting For Elderly Population In Malaysia: A Comparison Between Lee-Carter Model And Heligman-Pollard Model
The number of population in many countries might undergo dramatic changes in the coming decades due to continuous increases in life expectancy. The sustained reduction in mortality rates and its systematic underestimation has attracted a significant interest of worldwide researchers in recent yea...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | en_US |
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The number of population in many countries might undergo dramatic changes in the
coming decades due to continuous increases in life expectancy. The sustained
reduction in mortality rates and its systematic underestimation has attracted a
significant interest of worldwide researchers in recent years due to its potential impact
on population size and structure, social security systems, and the life insurance and
pensions industry (from an actuarial perspective). Among all projection methods, Lee-
Carter model has been widely accepted by the actuarial community. In addition,
Heligman-Pollard model is a well-known parametric model that has been widely used
by researchers. Therefore, this research explores the use of the Lee-Carter model and
Heligman-Pollard model to forecast the mortality rates for elderly population in
Malaysia. Since the model involves nonlinear equations that are explicitly difficult to
solve, the Matrix Laboratory Version 8.0 (MATLAB 8.0) software is used in the study
while Statistical Package for the Social Sciences Version 22.0 (SPSS 22.0) is applied
to forecast all the parameters according to Autoregressive Integrated Moving Average
(ARIMA) for both models. The empirical data sets of Malaysian population for period
of 1981 to 201 5 for both genders is considered, which the period of 198 1 to 2010 is
used as the "training set" and the period of 201 1 to 2015 as the ''testing set". In order
to investigate the accuracy of the estimation, the forecasted results are compared
against actual data of mortality rates using Mean Absolute Percentage Error ( W E ) .
The result shows that the Heligman-Pollard model fits mortality rates better for male
while Lee-Carter model performs better for female elderly population in Malaysia
Apart from that, the forecasting of mortality rates for the 10 upcoming years, from
2016 until year 2025 shows the continuation of the declining trends of mortality since
the past 35 years of study. |
|---|