Mortality and Life Expectancy Forecasting for a Group of Populations in Developed Countries: A Robust Multilevel Functional Data Method

Conference paper


A robust multilevel functional data method is proposed to forecast age-specific mortality rate and life expectancy for two or more populations in developed countries with high-quality vital registration systems. It uses a robust multilevel functional principal component analysis of aggregate and population-specific data to extract the common trend and population-specific residual trend among populations. This method is applied to age- and sex-specific mortality rate and life expectancy for the United Kingdom from 1922 to 2011, and its forecast accuracy is then further compared with standard multilevel functional data method. For forecasting both age-specific mortality and life expectancy, the robust multilevel functional data method produces more accurate point and interval forecasts than the standard multilevel functional data method in the presence of outliers.


Markov Chain Monte Carlo Forecast Error Prediction Interval Forecast Accuracy Functional Principal Component Analysis 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



The author is grateful for the invitation by Professor Graciela Boente to participate the ICORS2015 conference. The author thanks comments and suggestions received from the participants of the ICORS2015 conference, and the participants of the Bayesian methods for population estimation workshop held at the Australian Bureau of Statistics in May, 2015.


  1. Alkema L, Raftery AE, Gerland P, Clark SJ, Pelletier F, Buettner T, Heilig GK (2011) Probabilistic projections of the total fertility rate for all countries. Demography 48(3):815–839CrossRefGoogle Scholar
  2. Booth H (2006) Demographic forecasting: 1980–2005 in review. Int J Forecast 22(3):547–581CrossRefGoogle Scholar
  3. Booth H, Tickle L (2008) Mortality modelling and forecasting: a review of methods. Ann Actuarial Sci 3(1–2):3–43CrossRefGoogle Scholar
  4. Chiou JM (2012) Dynamical functional prediction and classification, with application to traffic flow prediction. Ann Appl Statis 6(4):1588–1614MathSciNetCrossRefMATHGoogle Scholar
  5. Crainiceanu CM, Goldsmith JA (2010) Bayesian functional data analysis using WinBUGS. J Statis Softw 32(11)Google Scholar
  6. Crainiceanu CM, Staicu AM, Di CZ (2009) Generalized multilevel functional regression. J Am Statis Assoc 104(488):1550–1561MathSciNetCrossRefMATHGoogle Scholar
  7. Delwarde A, Denuit M, Guillén M, Vidiella-i-Anguera A (2006) Application of the Poisson log-bilinear projection model to the G5 mortality experience. Belgian Actuarial Bull 6(1):54–68Google Scholar
  8. Di CZ, Crainiceanu CM, Caffo BS, Punjabi NM (2009) Multilevel functional principal component analysis. Ann Appl Statis 3(1):458–488MathSciNetCrossRefMATHGoogle Scholar
  9. Girosi F, King G (2008) Demographic forecasting. Princeton University Press, PrincetonGoogle Scholar
  10. Gneiting T, Katzfuss M (2014) Probabilistic forecasting. Ann Rev Statis Appl 1:125–151CrossRefGoogle Scholar
  11. Gneiting T, Raftery AE (2007) Strictly proper scoring rules, prediction and estimation. J Am Statis Assoc 102(477):359–378MathSciNetCrossRefMATHGoogle Scholar
  12. Greven S, Crainiceanu C, Caffo B, Reich D (2010) Longitudinal functional principal component analysis. Electron J Statis 4:1022–1054MathSciNetCrossRefMATHGoogle Scholar
  13. He X, Ng P (1999) COBS: qualitatively constrained smoothing via linear programming. Comput Statis 14:315–337CrossRefMATHGoogle Scholar
  14. Hubert M, Rousseeuw P, Verboven S (2002) A fast method for robust principal components with applications to chemometrics. Chemom Intell Lab Syst 60(1–2):101–111CrossRefGoogle Scholar
  15. Hubert M, Rousseeuw P, Branden K (2005) ROBPCA: a new approach to robust principal component analysis. Technometrics 47(1):64–79MathSciNetCrossRefGoogle Scholar
  16. Human Mortality Database (2015) University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). Accessed 8 March 2013.
  17. Hyndman RJ, Booth H (2008) Stochastic population forecasts using functional data models for mortality, fertility and migration. Int J Forecast 24(3):323–342CrossRefGoogle Scholar
  18. Hyndman RJ, Ullah MS (2007) Robust forecasting of mortality and fertility rates: a functional data approach. Comput Statis Data Anal 51(10):4942–4956MathSciNetCrossRefMATHGoogle Scholar
  19. Lee RD (2006) Mortality forecasts and linear life expectancy trends. In: Bengtsson T (ed) Perspectives on mortality forecasting, vol III. The linear rise in life expectancy: History and prospects, no. 3 in Social Insurance Studies, Swedish National Social Insurance Board, Stockholm, pp 19–39Google Scholar
  20. Lee RD, Carter LR (1992) Modeling and forecasting U.S. mortality. J Am Statis Assoc 87(419):659–671Google Scholar
  21. Li J (2013) A Poisson common factor model for projecting mortality and life expectancy jointly for females and males. Popul Stud 67(1):111–126CrossRefGoogle Scholar
  22. Li N, Lee R (2005) Coherent mortality forecasts for a group of population: an extension of the Lee-Carter method. Demography 42(3):575–594CrossRefGoogle Scholar
  23. Li N, Lee R, Gerland P (2013) Extending the Lee-Carter method to model the rotation of age patterns of mortality decline for long-term projections. Demography 50(6):2037–2051CrossRefGoogle Scholar
  24. Preston SH, Heuveline P, Guillot M (2001) Demography: measuring and modelling population process. Blackwell, OxfordGoogle Scholar
  25. Raftery AE, Li N, Ševčíková H, Gerland P, Heilig GK (2012) Bayesian probabilistic population projection for all countries. Proc Natl Acad Sci USA 109(35):13,915–13,921Google Scholar
  26. Raftery AE, Chunn JL, Gerland P, Ševčíková H (2013) Bayesian probabilistic projections of life expectancy for all countries. Demography 50(3):777–801CrossRefGoogle Scholar
  27. Raftery AE, Lalic N, Gerland P (2014) Joint probabilistic projection of female and male life expectancy. Demographic Res 30:795–822CrossRefGoogle Scholar
  28. Renshaw AE, Haberman S (2003) Lee-Carter mortality forecasting with age-specific enhancement. Insur.: Math Econ 33(2):255–272Google Scholar
  29. Rice J, Silverman B (1991) Estimating the mean and covariance structure nonparametrically when the data are curves. J R Statis Soc Ser B 53(1):233–243MathSciNetMATHGoogle Scholar
  30. Ševčiková H, Li N, Kantorová V, Gerland P, Raftery AE (2015) Age-specific mortality and fertility rates for probabilistic population projections. Working Paper, University of Washington.
  31. Shang HL (2016) Mortality and life expectancy forecasting for a group of populations in developed countries: a multilevel functional data method. Ann Appl Stat, (in press)Google Scholar
  32. Shang HL, Booth H, Hyndman RJ (2011) Point and interval forecasts of mortality rates and life expectancy: a comparison of ten principal component methods. Demographic Res 25(5):173–214CrossRefGoogle Scholar
  33. Tickle L, Booth H (2014) The longevity prospects of Australian seniors: an evaluation of forecast method and outcome. Asia-Pacific J Risk Insur 8(2):259–292Google Scholar
  34. Wiśniowski A, Smith PWF, Bijak J, Raymer J, Forster JJ (2015) Bayesian population forecasting: extending the Lee-Carter method. Demography 52(3):1035–1059CrossRefGoogle Scholar
  35. Yao F, Müller HG, Wang J (2005) Functional data analysis for sparse longitudinal data. J Am Statis Assoc 100(470):577–590MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer India 2016

Authors and Affiliations

  1. 1.Research School of Finance, Actuarial Studies and StatisticsAustralian National UniversityCanberraAustralia

Personalised recommendations