Page 128 - Contributed Paper Session (CPS) - Volume 3

CPS1965 Chin Tsung R. et al.
4. Conclusion
A major advantage of estimating mortality rates using cubic spline is that
it allows for age-specific mortality estimates compared to the original LC
model. This allows for more specific age group mortality estimation that is we
can estimate mortality rates for ages which are not the midpoint of an age
group. Additionally, compared to the LC Model which is not smooth, the cubic
spline estimate is continuous up to the second derivative. Another advantage
of the cubic spline method is the potentially lower number of model
parameters. In comparison, the number of parameters using the spline
approach would be determined by the number of knots and restrictions used.
One of the disadvantages in fitting a cubic spline to the mortality dataset of
Malaysia is the lack of granularity of the age groups as it limits the choice of
knots. However, as the estimated mortality models the error is relatively small,
it is reasonable to say that the model provides adequate estimates.

References
1. Hyndman, R.J. (2006). R package, demography: Forecasting Mortality,
Fertility, Migration and Population Data.
2. Lee, R. D., & Carter, L. R. (1992). Modeling and forecasting US mortality.
Journal of the American Statistical Association, 87(419), 659-671.
3. Zhao, B. B. (2012). A modified Lee–Carter model for analysing short-
base-period data. Population studies, 66(1), 39-52.

117 | I S I W S C 2 0 1 9