| The traditional methodologies to forecast mortality are based on the assumption that the life expectancy has an upper limit or the life expectancy increase is limited. Lee Ronald D. and Carter Lawrence R. (1992) proposed a method for forecasting future mortality. Several advantages are claimed: a parsimonious demographic model is combined with statistical time-series methods, the method involves no subjective 'judgements'. According to the empirical data of mortality rates, Lee-Carter method is employed to forecast the mortality rates. The log-additive model and the ARIMA model are two main features of Lee-Carter model. Demographers in the developed countries are interested in the Lee-Carter method and like to use it.This dissertation presents an interpretation of the model underpinning the Lee-Carter methodology for forecasting mortality rates and other vital rates in China. There's an analysis of the age-specific mortality rates for males and females in China during the period 1986-2002 and through a consideration of the forecasts generated by the Lee-Carter method. This dissertation also compares different approaches to forecast life expectancies for each gender, and then the forecasts under the Lee-Carter are compared with the official projections from the National Bureau of Statistics of China and the United Nations Population Division.The main purpose of this dissertation is to improve the methodology for forecasting Chinese mortality in order to enhance model performance and increase forecasting power. This dissertation examines the way in which the Lee-Carter model can be introduced into China with the limited mortality data available. Two questions will be answered in this dissertation. The first is how to apply the Lee-Carter method and the GLM method to the Chinese data, in which the historical data is only available for a small number of time periods. The second question relates to the quality of results that can be derived through applying the Lee-Carter method. |
PDF Full Text Request