×

A group regularisation approach for constructing generalised age-period-cohort mortality projection models. (English) Zbl 1484.91405

Summary: Given the rapid reductions in human mortality observed over recent decades and the uncertainty associated with their future evolution, there have been a large number of mortality projection models proposed by actuaries and demographers in recent years. Many of these, however, suffer from being overly complex, thereby producing spurious forecasts, particularly over long horizons and for small, noisy data sets. In this paper, we exploit statistical learning tools, namely group regularisation and cross-validation, to provide a robust framework to construct discrete-time mortality models by automatically selecting the most appropriate functions to best describe and forecast particular data sets. Most importantly, this approach produces bespoke models using a trade-off between complexity (to draw as much insight as possible from limited data sets) and parsimony (to prevent over-fitting to noise), with this trade-off designed to have specific regard to the forecasting horizon of interest. This is illustrated using both empirical data from the Human Mortality Database and simulated data, using code that has been made available within a user-friendly open-source R package StMoMo.

MSC:

91G05 Actuarial mathematics
91D20 Mathematical geography and demography

Software:

Demography; R; StMoMo; ISLR
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Aro, H. and Pennanen, T. (2011) A user-friendly approach to stochastic mortality modelling. European Actuarial Journal, 1(2), 151-167.
[2] Atance, D. and Debón, A. (2020) A comparison of forecasting mortality models using resampling methodsMathematics, 8(9), 1550. https://doi.org/10.3390/math8091550.
[3] Barigou, K., Loisel, S. and Salhi, Y. (2021) Parsimonious predictive mortality modeling by regularization and cross-validation with and without Covid-type effect. Risks, 9(1), 5.
[4] Bergmeir, C. and Bentez, J.M. (2012) On the use of cross-validation for time series predictor evaluation. Information Sciences, 191, 192-213. https://doi.org/10.1016/j.ins.2011.12.028.
[5] Bergmeir, C., Costantini, M. and Bentez, J.M. (2014) On the usefulness of cross-validation for directional forecast evaluation. Computational Statistics and Data Analysis, 76, 132-143. https://doi.org/10.1016/j.csda.2014.02.001. · Zbl 06983969
[6] Bergmeir, C., Hyndman, R. and Koo, B. (2018) A note on the validity of cross-validation for evaluating autoregressive time series prediction. Computational Statistics and Data Analysis, 120, 70-83. https://doi.org/10.1016/j.csda.2017.11.003. · Zbl 1469.62021
[7] Biffis, E. (2005) Affine processes for dynamic mortality and actuarial valuations. Insurance: Mathematics and Economics, 37(3), 443-468. https://doi.org/10.1016/j.insmatheco.2005.05.003. · Zbl 1129.91024
[8] Booth, H., Hyndman, R.J., Tickle, L. and De Jong, P. (2006) Lee-Carter mortality forecasting: A multi-country comparison of variants and extensions. Demography, 15, 289-310. https://doi.org/10.4054/DemRes.2006.15.9.
[9] Booth, H. and Tickle, L. (2008) Mortality modelling and forecasting: A review of methods. Annals of Actuarial Science, 31(1-2), 3-43. https://doi.org/10.1017/S1748499500000440.
[10] Breheny, P. and Huang, J. (2015) Group descent algorithms for nonconvex penalized linear and logistic regression models with grouped predictors. Statistics and Computing, 25(2), 173-87. https://doi.org/10.1007/s11222-013-9424-2. · Zbl 1331.62359
[11] Brouhns, N., Denuit, M. and Vermunt, J. (2002a) A Poisson log-linear regression approach to the construction of projected life tables. Insurance: Mathematics and Economics, 31(3), 373-393. · Zbl 1074.62524
[12] Brouhns, N., Denuit, M. and Vermunt, J.K. (2002b) Measuring the longevity risk in mortality projections. Bulletin of the Swiss Association of Actuaries, 2(1), 105-130. · Zbl 1187.62158
[13] Cairns, A., Blake, D. and Dowd, K. (2006) A two-factor model for stochastic mortality with parameter uncertainty: Theory and calibration. The Journal of Risk and Insurance, 73(4), 687-718.
[14] Cairns, A., Blake, D., Dowd, K., Coughlan, G., Epstein, D. and Khalaf-Allah, M. (2011) Mortality density forecasts: An analysis of six stochastic mortality models. Insurance: Mathematics and Economics, 48(3), 355-367. https://doi.org/10.1016/j.insmatheco.2010.12.005. · Zbl 1231.91179
[15] Cairns, A., Blake, D., Dowd, K., Coughlan, G., Epstein, D., Ong, A. and Balevich, I. (2010) A quantitative comparison of stochastic mortality models using data from England and Wales and the United States. North American Actuarial Journal, 13(1), 1-35. · Zbl 1484.91376
[16] Cairns, A.J.G., Blake, D., Dowd, K., Coughlan, G.D., Epstein, D. and Khalaf-Allah, M. (2008) Mortality density forecasts: An analysis of six stochastic mortality models. Pensions Institute Discussion Paper PI-0801. http://papers.ssrn.com/sol3/papers.cfm?abstract
[17] Cairns, A.J.G., Blake, D., Dowd, K., Coughlan, G.D., Epstein, D. and Khalaf-Allah, M. (2011) Mortality density forecasts: An analysis of six stochastic mortality models. Insurance: Mathematics and Economics, 48(3), 355-367. · Zbl 1231.91179
[18] Currie, I. (2006) Smoothing and forecasting mortality rates with p-splines. https://www.macs.-hw.ac.uk/Iain.
[19] Currie, I. (2016) On fitting generalized linear and non-linear models of mortality. Scandinavian Actuarial Journal, (4), 356-383. https://doi.org/10.1080/03461238.2014.928230. · Zbl 1401.91123
[20] Currie, I. (2020) Constraints, the identifiability problem and the forecasting of mortality. Annals of Actuarial Science, 14(2), 537-566. https://doi.org/10.1017/S1748499520000020.
[21] Currie, I., Durban, M. and Eilers, P. (2004) Smoothing and forecasting mortality rates. Statistical Modelling, 4(4), 279-298. · Zbl 1061.62171
[22] Delwarde, A., Denuit, M. and Eilers, P. (2007) Smoothing the Lee-Carter and Poisson log-bilinear models for mortality forecasting: A penalized log-likelihood approach. Statistical Modelling, 7(1), 29-48. https://doi.org/10.1177/1471082X0600700103. · Zbl 07257671
[23] Devriendt, S., Antonio, K., Reynkens, T. and Verbelen, R. (2020) Sparse regression with multi-type regularized feature modeling. Insurance: Mathematics and Economics, 96, 248-261. · Zbl 1460.91218
[24] Dowd, K., Cairns, A., Blake, D., Coughlan, G., Epstein, D. and Khalaf-Allah, M. (2009) Backtesting stochastic mortality models: An ex post evaluation of multiperiod-ahead density forecasts. North American Actuarial Journal, 14(3), 281-98.
[25] Dowd, K., Cairns, A.J.G. and Blake, D. (2020) CBDX: A workhorse mortality model from the Cairns-Blake-Dowd family. Annals of Actuarial Science, 14(2), 445-460. https://doi.org/10.1017/s1748499520000159.
[26] Fan, J. and Li, R. (2001) Variable selection via nonconcave penalized likelihood and its oracle properties. Journal of the American Statistical Association, 96(456), 1348-1360. https://doi.org/10.1198/016214501753382273. · Zbl 1073.62547
[27] Gneiting, T. and Raftery, A. (2007) Strictly proper scoring rules, prediction, and estimation. Journal of the American Statistical Association, 102(477), 359-78. · Zbl 1284.62093
[28] Green, K. and Scott Armstrong, J. (2015) Simple versus complex forecasting: The evidence. Journal of Business Research, 68(8), 1678-1685. https://doi.org/10.1016/j.jbusres.2015.03.026.
[29] Guibert, Q., Lopez, O. and Piette, P. (2019) Forecasting mortality rate improvements with a high-dimensional VAR. Insurance: Mathematics and Economics, 88, 255-72. https://doi.org/10.1016/j.insmatheco.2019.07.004. · Zbl 1425.91223
[30] Haberman, S. and Renshaw, A. (2011) A comparative study of parametric mortality projection models. Insurance: Mathematics and Economics, 48(1), 35-55. https://doi.org/10.1016/j.insmatheco.2010.09.003.
[31] Hainaut, D. and Denuit, M. (2020) Wavelet-based feature extraction for mortality projection. ASTIN Bulletin, 50(3), 675-707. https://doi.org/10.1017/asb.2020.18. · Zbl 1454.91190
[32] Hastie, T., Tibshirani, R. and Friedman, J. (2001) The Elements of Statistical Learning. https://doi.org/10.1007/b94608. · Zbl 0973.62007
[33] Holford, T.R. (1983) The estimation of age, period and cohort effects for vital rates. Biometrics, 331-324. https://doi.org/10.2307/2531004.
[34] . (2020) University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). www.mortality.org.
[35] Hunt, A. and Blake, D. (2014) A General procedure for constructing mortality models. North American Actuarial Journal, 18(1), 116-38. https://doi.org/10.1080/10920277.2013.852963. · Zbl 1412.91045
[36] Hunt, A. and Blake, D. (2020a) Identifiability in age/period/cohort mortality models. Annals of Actuarial Science, 14(2), 550-536. https://doi.org/10.1017/S1748499520000123.
[37] Hunt, A. and Blake, D. (2020b) Identifiability in age/period mortality models. Annals of Actuarial Science, 14(2), 461-499. https://doi.org/10.1017/s1748499520000111.
[38] Hunt, A. and Blake, D. (2021a) A Bayesian approach to modeling and projecting cohort effects. North American Actuarial Journal, 25(sup1), S235-S254. https://doi.org/10.1080/10920277.2019.1649157. · Zbl 1461.91245
[39] Hunt, A. and Blake, D. (2021b) On the structure and classification of mortality models. North American Actuarial Journal, 25(sup1), S215-S234. https://doi.org/10.1080/10920277.2019.1649156. · Zbl 1461.91244
[40] Hunt, A. and Villegas, A.M. (2015) Robustness and convergence in the Lee-Carter model with cohorts. Insurance: Mathematics and Economics, 64, 186-202. · Zbl 1348.62241
[41] Hyndman, R., Booth, H., Tickle, L. and Maindonald, J. (2017) Package ‘demography’. http://cran.r-project.org/package=demogra.
[42] James, G., Witen, D., Hastie, T. and Tibshirani, R. (2013) An Introduction to Statistical Learning with Applications in R. New York: Springer.
[43] Karanikolos, M., Leon, D.A., Smith, P.C. and Mckee, M. (2012) Minding the gap: Changes in life expectancy in the Baltic States compared with Finland. Journal of Epidemiology and Community Health, 66(11), 1043-1049. https://doi.org/10.1136/jech-2011-200879.
[44] Lee, R. and Carter, L. (1992) Modeling and forecasting U.S. Mortality. Journal of the American Statistical Association, 87(419), 673-674. · Zbl 1351.62186
[45] Li, H. and O’Hare, C. (2017) Semi-parametric extensions of the Cairns-Blake-Dowd model: A one-dimensional kernel smoothing approach. Insurance: Mathematics and Economics, 77, 166-176. · Zbl 1397.91290
[46] Li, H. and Shi, Y. (2021) Mortality forecasting with an age-coherent sparse VAR model. Risks, 9(35). https://doi.org/10.1017/asb.2020.39.
[47] Li, J.S.H., Zhou, R., Liu, Y., Graziani, G., Dale Hall, R., Haid, J., Peterson, A. and Pinzur, L. (2020) Drivers of mortality dynamics: Identifying age/period/cohort components of historical U.S. mortality improvements. North American Actuarial Journal, 24(2), 228-50. https://doi.org/10.1080/10920277.2020.1716808. · Zbl 1454.91199
[48] Li, J.S.H., Zhou, R. and Hardy, M. (2015) A step-by-step guide to building two-population stochastic mortality models. Insurance: Mathematics and Economics, 63, 121-34. https://doi.org/10.1016/j.insmatheco.2015.03.021. · Zbl 1348.91164
[49] Macdonald, A.S., Richards, S.J. and Currie, I.D. (2018) Modelling Mortality with Actuarial Applications. Cambridge: Cambridge University Press. · Zbl 1410.62005
[50] Njenga, C.N. and Sherris, M. (2011) Longevity risk and the econometric analysis of mortality trends and volatility. Asia-Pacific Journal of Risk and Insurance, 5(2). https://doi.org/10.2202/2153-3792.1115.
[51] Plat, R. (2009) On stochastic mortality modeling. Insurance: Mathematics and Economics, 45(3), 393-404. https://doi.org/10.1016/j.insmatheco.2009.08.006. · Zbl 1231.91227
[52] Racine, J. (2000) Consistent cross-validatory model-selection for dependent data: hv-block cross-validation. Journal of Econometrics, 99(1), 39-61. https://doi.org/10.1016/S0304-4076(00)00030-0. · Zbl 1011.62118
[53] Renshaw, A. and Haberman, S. (2003a) Lee-Carter mortality forecasting with age-specific enhancement. Insurance: Mathematics and Economics, 33(2), 255-72. · Zbl 1103.91371
[54] Renshaw, A. and Haberman, S. (2003b) On the forecasting of mortality reduction factors. Insurance: Mathematics and Economics, 32(3), 379-401. https://doi.org/10.1016/S0167-6687(03)00118-5. · Zbl 1025.62041
[55] Renshaw, A. and Haberman, S. (2006) A cohort-based extension to the Lee-Carter model for mortality reduction factors. Insurance: Mathematics and Economics, 38(3), 556-570. https://doi.org/10.1016/j.insmatheco.2005.12.001. · Zbl 1168.91418
[56] Tashman, L.J. (2000) Out-of-sample tests of forecasting accuracy: An analysis and review. International Journal of Forecasting, 16(4), 437-50. https://doi.org/10.1016/S0169-2070(00)00065-0.
[57] Tibshirani, R. (1996) Regression selection and shrinkage via the lasso. Journal of the Royal Statistical Society, 58(1), 267-288. https://doi.org/10.2307/2346178. · Zbl 0850.62538
[58] Vandekerckhove, J., Matzke, D. and Wagenmakers, E.-J. (2015). Model comparison and the principle of parsimony. In The Oxford Handbook of Computational and Mathematical Psychology.https://doi.org/10.1093/oxfordhb/9780199957996.013.14.
[59] Venter, G. and Sahin, S. (2018) Parsimonious parameterization of age-period-cohort models by Bayesian shrinkage. ASTIN Bulletin, 48(1), 89-110. https://doi.org/10.1017/asb.2017.21. · Zbl 1390.62226
[60] Villegas, A., Kaishev, V. and Millossovich, P. (2018) StMoMo: An R package for stochastic mortality modeling. Journal of Statistical Software84(3). https://doi.org/10.18637/jss.v084.i03.
[61] Yuan, M. and Lin, Y. (2006) Model selection and estimation in regression with grouped variables. Journal of the Royal Statistical Society. Series B: Statistical Methodology, 68(1), 49-67. https://doi.org/10.1111/j.1467-9868.2005.00532.x. · Zbl 1141.62030
[62] Zhang, C.H. (2010) Nearly unbiased variable selection under minimax concave penalty. Annals of Statistics, 38(2), 894-942. https://doi.org/10.1214/09-AOS729. · Zbl 1183.62120
[63] Zou, H. and Hastie, T. (2005) Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society, 67(2), 301-20. https://doi.org/10.1111/j.1467-9868.2005.00503.x. · Zbl 1069.62054
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.