Mortality Forecasting Using Lee Carter Model Implemented to French Mortality Data

Abstract

For the reason of simplicity the Lee and Carter (LC) method is getting widely adopted for long-run forecasts of age specific mortality rates. In
this paper the LC model is applied to French mortality data to demonstrate the mortality results. The age-specific death rates are used for the
period 1816 to 2006. The index of the level of mortality, and the shape and sensitivity coefficients for each age are obtained through the LC
method. The autoregressive moving average and the singular value decomposition models are used to forecast the general index for a long
period of time that goes from 2007 to 2056. The projection is useful since the projected mortality rates can be used to project life expectancy at
birth which is the widely used social indicator in demography.