Five different distributions for the Lee-Carter model of mortality forecasting: a comparison using GAS models. (English) Zbl 1394.91327

Summary: This paper extends the well-known Lee-Carter model used for forecasting mortality rates by utilizing a new class of time series models, known as Generalized Autoregressive Score (GAS) or Dynamic Conditional Score (DCS) models. This framework can be used to derive a wide range of non-Gaussian time series models with time varying coefficients and has shown to be very successful in financial applications. In this paper we propose five probability models (Poisson, binomial, negative binomial, Gaussian and beta) based on the GAS framework to estimate the Lee-Carter parameters and dynamically forecast the mortality rates using a single unified step. The models are applied to the mortality rates time series for the male population of the United States, Sweden, Japan and the UK. Diagnostic tests are performed on quantile residuals, model selection is made via AIC and predictive accuracy of the models is compared using the Diebold-Mariano test. We conclude that, amongst the proposed models, the negative binomial extension of the Lee-Carter model is the most appropriate for forecasting mortality rates.


91D20 Mathematical geography and demography
62P05 Applications of statistics to actuarial sciences and financial mathematics
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
91B30 Risk theory, insurance (MSC2010)
Full Text: DOI


[1] Berndt, E. R.; Hall, B. H.; Hall, R. E.; Hausman, J. A., Estimation and inference in nonlinear structural models, Ann. Econ. Soc. Meas., 3, 4, 653-665, (1974)
[2] Blasques, F.; Koopman, S. J.; Lucas, A., Information-theoretic optimality of observation-driven time series models for continuous responses, Biometrika, 102, 2, (2015) · Zbl 1452.62620
[3] Bollerslev, T., Generalized autoregressive conditional heteroskedasticity, J. Econometrics, 31, 3, 307-327, (1986) · Zbl 0616.62119
[4] Brouhns, N.; Denuit, M.; Vermunt, J. K., A Poisson log-bilinear regression approach to the construction of projected lifetables, Insurance Math. Econom., 31, 3, 373-393, (2002) · Zbl 1074.62524
[5] Broyden, C. G., The convergence of a class of double-rank minimization algorithms, J. Inst. Math. Appl., 6, 1, 76-90, (1970) · Zbl 0223.65023
[6] Chen, H., MacMinn, R.D., Sun, T., 2014. Mortality Dependence and Longevity Bond Pricing: A Dynamic Factor Copula Mortality Model with the GAS Structure - http://www.gasmodel.com/gaspapers.htm.
[7] Cossette, H.; Delwarde, A.; Denuit, M.; Guillot, F.; Marceau, É., Pension plan valuation and mortality projection: a case study with mortality data, N. Am. Actuar. J., 11, 2, 1-34, (2007)
[8] Creal, D., Koopman, S.J., Lucas, A., 2008. A General Framework for Observation Driven Time-Varying Parameter Models. Tinbergen Institute Discussion, 108/4.
[9] Creal, D.; Koopman, S. J.; Lucas, A., Generalized autoregressive score models with applications, J. Appl. Econometrics, 28, 5, 777-795, (2013)
[10] Creal, D.; Schwaab, B.; Koopman, S. J.; Lucas, A., Observation-driven mixed-measurement dynamic factor models with an application to credit risk, Rev. Econ. Stat., 96, 5, 898-915, (2014)
[11] Davis, R. A.; Dunsmuir, W. T.M.; Streett, S., Observation driven models for Poisson counts, Biometrika, 90, 4, 777-790, (2003) · Zbl 1436.62418
[12] De Jong, P.; Tickle, L., Extending Lee-Carter mortality forecasting, Math. Popul. Stud., 13, 1, 1-18, (2006) · Zbl 1151.91742
[13] Delwarde, A.; Denuit, M.; Partrat, C., Negative binomial version of the Lee-Carter model for mortality forecasting, Appl. Stoch. Models Bus. Ind., 23, 5, 385-401, (2007) · Zbl 1150.91426
[14] Diebold, F. X.; Mariano, R. S., Comparing predictive accuracy, J. Bus. Econom. Statist., 20, 1, (2002)
[15] Engle, R. F., Autoregressive conditional heteroscedasticity with estimates of the variance of united kingdom inflation, Econometrica, 50, 4, 987-1007, (1982) · Zbl 0491.62099
[16] Engle, R. F.; Russell, J. R., Autoregressive conditional duration: a new model for irregularly spaced transaction data, Econometrica, 66, 5, 1127-1162, (1998) · Zbl 1055.62571
[17] Fletcher, R., A new approach to variable metric algorithms, Comput. J., 13, 3, 317-322, (1970) · Zbl 0207.17402
[18] Goldfarb, D., A family of variable metric updates derived by variational means, Math. Comp., 24, 109, 23-26, (1970) · Zbl 0196.18002
[19] Haberman, S., Renshaw, A., 2008. On simulation-based approaches to risk measurement in mortality with specific reference to binomial Lee-Carter modelling. In: Society of Actuaries Living to 100 Symposium. · Zbl 1152.91598
[20] Harvey, A. C., Dynamic models for volatility and heavy tails: with applications to financial and economic time series (no. 52), (2013), Cambridge University Press · Zbl 1326.62001
[21] Lee, R. D.; Carter, L., Modeling and forecasting the time series of US mortality, J. Amer. Statist. Assoc., 87, 659, 71, (1992)
[22] Pitacco, E.; Denuit, M.; Haberman, S., Modelling longevity dynamics for pensions and annuity business, (2009), Oxford University Press · Zbl 1163.91005
[23] Renshaw, A. E.; Haberman, S., A cohort-based extension to the Lee-Carter model for mortality reduction factors, Insurance Math. Econom., 38, 3, 556-570, (2006) · Zbl 1168.91418
[24] Russell, J.R., 2001. Econometric modeling of multivariate irregularly-spaced high-frequency data. Unpublished manuscript, University of Chicago, Graduate School of Business.
[25] Shanno, D. F., Conditioning of quasi-Newton methods for function minimization, Mathematics of Computation, 24, 111, 647-656, (1970) · Zbl 0225.65073
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.