臺灣大學數學系專題演講
NTU Math Colloquium
 

演講者：Professor Han-Lin Shang  (Australian National University)

講  題：Mortality and life expectancy forecasting for a group of populations in developed countries: a multilevel functional data method

時  間：2013年12月29日 (星期一) 14:00~15:00

地  點：臺灣大學天數館202室

摘  要：Individual forecasts of mortality for multiple populations often diverge in the long run. A multilevel functional data method is adapted for forecasting age-specific mortality rates for two or more populations in developed countries with high quality vital registration systems. It uses multilevel functional principal component analysis of aggregated data and individual data to extract the common trend and individual trend among populations. If the forecasts of individual trends do not show a long-term trend, then convergence in forecasts is achieved. This method is first applied to age- and sex-specific data for the United Kingdom, and its forecast accuracy is then further compared with several existing methods, including independent functional data and product-ratio methods, through a multi-country comparison. The proposed method is also demonstrated by age- and sex- and state-specific data in Australia, where the convergence in forecasts is achieved by both state and sex. For forecasting mortality and life expectancy, the multilevel functional data method produces slightly better forecast accuracy to the product-ratio method, and they are both more accurate than the independent functional data method in overall terms.