Estimating parameters of Makeham's law of mortality

Abstract

Actuaries and demographers have a long tradition of using collateral data to improve mortality estimates. Two main approaches have been used to accomplish the improvement, mortality laws and model life tables. The most common mortality law that used to illustrate t h e life tables is Makeham's law. In Sri Lanka, Census and Statistics Department presents t h e life table and calculation of life table had been done using software package Mortpak-lite developed by united nations and the most insurance companies use UK standard life tables to construct insurance premiums. So my purpose is t o construct t h e model life t a b l e for Sri Lanka, estimates t h e constants of Makeham's law of mortality that used to construct life table, and introduces a regression model that incorporates the early stages (below 20) and also wants to determine whether t h e r e ' s big difference for males and females. The Makeham's estimates are based upon the equation pi(x) = A + BC. Two methods are used t o estimates t h e parameters, namely least square method and using t h e definition of age specific death rate. The model is demonstrated t h e mortality data (year 2001) of Sri Lanka.