Bayesian model comparison for mortality forecasting
Bayesian model comparison for mortality forecasting
Stochastic models are appealing for mortality forecasting in their ability to generate intervals that quantify uncertainties underlying the forecasts. We present a fully Bayesian implementation of the age-period-cohort-improvement (APCI) model with overdispersion, which is compared with the Lee–Carter model with cohorts. We show that naive prior specification can yield misleading inferences, where we propose Laplace prior as an elegant solution. We also perform model averaging to incorporate model uncertainty. Our findings indicate that the APCI model offers better fit and forecast for England and Wales data spanning 1961–2002. Our approach also allows coherent inclusion of multiple sources of uncertainty, producing well-calibrated probabilistic intervals.
566-586
Wong, Jackie S.T.
22e8b168-4c09-4a35-a3b1-0398636566b2
Forster, Jonathan J.
8863fb7a-c11c-4821-b78c-4f4d278752bc
Smith, Peter W.F.
961a01a3-bf4c-43ca-9599-5be4fd5d3940
June 2023
Wong, Jackie S.T.
22e8b168-4c09-4a35-a3b1-0398636566b2
Forster, Jonathan J.
8863fb7a-c11c-4821-b78c-4f4d278752bc
Smith, Peter W.F.
961a01a3-bf4c-43ca-9599-5be4fd5d3940
Wong, Jackie S.T., Forster, Jonathan J. and Smith, Peter W.F.
(2023)
Bayesian model comparison for mortality forecasting.
Journal of the Royal Statistical Society, Series C (Applied Statistics), 72 (3), , [qlad021].
(doi:10.1093/jrsssc/qlad021).
Abstract
Stochastic models are appealing for mortality forecasting in their ability to generate intervals that quantify uncertainties underlying the forecasts. We present a fully Bayesian implementation of the age-period-cohort-improvement (APCI) model with overdispersion, which is compared with the Lee–Carter model with cohorts. We show that naive prior specification can yield misleading inferences, where we propose Laplace prior as an elegant solution. We also perform model averaging to incorporate model uncertainty. Our findings indicate that the APCI model offers better fit and forecast for England and Wales data spanning 1961–2002. Our approach also allows coherent inclusion of multiple sources of uncertainty, producing well-calibrated probabilistic intervals.
Text
qlad021
- Version of Record
More information
Accepted/In Press date: 1 February 2023
e-pub ahead of print date: 22 March 2023
Published date: June 2023
Identifiers
Local EPrints ID: 476901
URI: http://eprints.soton.ac.uk/id/eprint/476901
ISSN: 0035-9254
PURE UUID: 2469e4ea-2eb9-4d38-83b4-ebf6a48fc821
Catalogue record
Date deposited: 19 May 2023 16:30
Last modified: 17 Mar 2024 02:37
Export record
Altmetrics
Contributors
Author:
Jackie S.T. Wong
Author:
Jonathan J. Forster
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
