User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

A Neural-Network Analyzer for Mortality Forecast

Primary tabs

download
  • Open access
  • PDF
  • 618.57 K
Hainaut, Donatien [UCL]

This article proposes a neural-network approach to predict and simulate human mortality rates. This semi-parametric model is capable to detect and duplicate non-linearities observed in the evolution of log-forces of mortality. The method proceeds in two steps. During the first stage, a neural-network-based generalization of the principal component analysis summarizes the information carried by the surface of log-mortality rates in a small number of latent factors. In the second step, these latent factors are forecast with an econometric model. The term structure of log-forces of mortality is next reconstructed by an inverse transformation. The neural analyzer is adjusted to French, UK and US mortality rates, over the period 1946–2000 and validated with data from 2001 to 2014. Numerical experiments reveal that the neural approach has an excellent predictive power, compared to the Lee–Carter model with and without cohort effects.

Document type Article de périodique (Journal article) – Article de recherche
Access type Accès libre
Publication date 2018
Language Anglais
Journal information "ASTIN Bulletin" - Vol. 48, no. 2, p. 481-508 (2018)
Peer reviewed yes
Publisher Cambridge University Press (CUP)
issn 0515-0361
e-issn 1783-1350
Publication status Publié
Affiliation UCL - SSH/LIDAM/ISBA - Institut de Statistique, Biostatistique et Sciences Actuarielles
Keywords Economics and Econometrics ; Accounting ; Finance
Links
  1. Dutse Journal of Pure and Applied Sciences, 1, 55 (2015)
  2. Antonio Katrien, Bardoutsos Anastasios, Ouburg Wilbert, Bayesian Poisson log-bilinear models for mortality projections with multiple populations, 10.1007/s13385-015-0115-6
  3. Insurance: Mathematics and Economics, 31, 373 (2002)
  4. Scandinavian Actuarial Journal, 2–3, 79 (2008)
  5. Scandinavian Actuarial Journal, 4, 356 (2016)
  6. Journal of the Royal Statistical Society, Series B., 34, 187 (1972)
  7. Cybenko G., Approximation by superpositions of a sigmoidal function, 10.1007/bf02551274
  8. Insurance: Mathematics and Economics, 53, 464 (2013)
  9. Dong D., McAvoy T.J., Nonlinear principal component analysis—Based on principal curves and neural networks, 10.1016/0098-1354(95)00003-k
  10. Biological Cybernetics, 77, 282 (1997)
  11. Insurance: Mathematics and Economics, 5, 236 (2012)
  12. Hornik Kurt, Approximation capabilities of multilayer feedforward networks, 10.1016/0893-6080(91)90009-t
  13. Soviet Physics Crystallography, 24, 519 (1979)
  14. Kramer Mark A., Nonlinear principal component analysis using autoassociative neural networks, 10.1002/aic.690370209
  15. Journal of the American Statistical Association, 87, 659 (1992)
  16. Lee Ronald, The Lee-Carter Method for Forecasting Mortality, with Various Extensions and Applications, 10.1080/10920277.2000.10595882
  17. Malthouse E.C., Limitations of nonlinear PCA as performed with generic neural networks, 10.1109/72.655038
  18. Neural Networks in Finance: Gaining Predictive Edge in the Market (2005)
  19. O’Hare Colin, Li Youwei, Explaining young mortality, 10.1016/j.insmatheco.2011.09.005
  20. Insurance: Mathematics and Economics, 35, 279 (2004)
  21. Modeling Longevity Dynamics for Pensions and Annuity Business (2009)
  22. Puddu Paolo Emilio, Menotti Alessandro, Artificial neural network versus multiple logistic function to predict 25-year coronary heart disease mortality in the Seven Countries Study, 10.1097/hjr.0b013e32832d49e1
  23. Puddu Paolo Emilio, Menotti Alessandro, Artificial neural networks versus proportional hazards Cox models to predict 45-year all-cause mortality in the Italian Rural Areas of the Seven Countries Study, 10.1186/1471-2288-12-100
  24. Puddu Paolo Emilio, Piras Paolo, Menotti Alessandro, Lifetime competing risks between coronary heart disease mortality and other causes of death during 50 years of follow-up, 10.1016/j.ijcard.2016.11.157
  25. Insurance: Mathematics and Economics, 33, 255 (2003)
  26. Insurance: Mathematics and Economics, 38, 556 (2006)
  27. van Berkum Frank, Antonio Katrien, Vellekoop Michel, The impact of multiple structural changes on mortality predictions, 10.1080/03461238.2014.987807
  28. Wong-Fupuy Carlos, Haberman Steven, Projecting Mortality Trends : Recent Developments in the United Kingdom and the United States, 10.1080/10920277.2004.10596137
  29. Insurance: Mathematics and Economics, 46, 254 (2010)
Bibliographic reference Hainaut, Donatien. A Neural-Network Analyzer for Mortality Forecast. In: ASTIN Bulletin, Vol. 48, no. 2, p. 481-508 (2018)
Permanent URL http://hdl.handle.net/2078.1/196618