Skip to Content

Li Ka Ki, Jackie

  • Paper Title: Mortality Projection Using the Logistic Model
  • Department Affiliation: Marketing and Management
  • Supervisors’ Names:
    • Doctor Nick Parr, 
    • Associate Professor Leonie Tickle


In this project, the logistic model is fitted to the observed mortality rates for each calendar year, and the estimated parameters are then projected into the future in order to sample a distribution for future mortality rates.       


This work attempts to project the parameters of the logistic model and focuses on the age range of 60 to 89. As the model for this age range is largely linear and there are only two parameters involved, projection is much more feasible than earlier work in the literature of using too many parameters. 

Key literature / theoretical perspective:

A. R. Thatcher. 1999. The long-term pattern of adult mortality and the highest attained age. Journal of the Royal Statistical Society Series A, 162: 5-43. 


The new approach is tested on several countries' mortality data and out-of-sample tests are conducted.         


The out-of-sample tests indicate that the forecasting performance of the new approach is satisfactory.  

Research limitations/implications:

As rising longevity becomes a significant issue for pension and annuity providers, the new approach in this project is potentially useful for projecting future mortality rates and pricing pensions and annuities.

Practical and Social implications:

The new approach provides new insights into projecting the mortality rates for those who are of retirement age. 


Logistic model, mortality rates, longevity