Overview
In the first part of this project we aim to produce the latest official decennial life table, ELT17, for the Office for National Statistics (ONS). This life table is based on the mortality experience of the England and Wales population during the years 2010, 2011 and 2012. We develop a methodology for smoothing UK mortality data incorporating uncertainty into the old-age mortality estimates.
The second part of the project involves a review of the current practice of setting the mortality assumptions in the UK National Population Projections carried out at the ONS. In this review we answer several methodological questions, including how to extrapolate mortality rates to the oldest ages, reduce the subjective assessment and correction, and project the future mortality improvement rates in the UK. Besides reviewing the ONS methodology and offering practical recommendations, we propose a comprehensive, coherent and flexible mortality projection method, which preserves the advantages of their current approach, and eliminates some of its limitations.
Publications & Activities
Forecasting Migration: Methods, Models, Uncertainty
Parliamentary Office for Science and Technology webinar (2020). (London via weblink)
Authors: Bijak J,
Modelling frontier mortality using Bayesian generalised additive models.
Journal of Official Statistics (2020). 37 (3) 569-589
Authors: Hilton J, Dodd E, Smith P, Forster J,
Stochastic modelling and projection of mortality improvements using a hybrid parametric/semi-parametric age-period-cohort model
Scandinavian Actuarial Journal (2020). 2021 (2) 134-155
Authors: Dodd E, Forster J, Bijak J, Smith P,
Modelling frontier mortality using Bayesian generalised additive models
United Nations Economic Commission for Europe Conference of European Statisticians (2019). (Belgrade, Serbia)
Authors: Hilton J, Dodd E, Forster J, Smith P,
Projecting UK Mortality using Bayesian Generalised Additive Models
Journal of the Royal Statistical Society (2018). 1-21
Authors: Hilton J, Dodd E, Forster J, Smith P,
Smoothing mortality data: The English Life Table, 2010-12
Journal of the Royal Statistical Society - Series A (2018). 181 (3) 717-735
Authors: Dodd E, Forster J, Bijak J, Smith P,
Prediction of settlement delay in critical illness insurance claims by using the generalized beta of the second kind distribution
Journal of the Royal Statistical Society Series C (2017). 66 (2) 273-294
Authors: Dodd E, Strefaris G,
Implementing a model-based approach to estimating mortality rates: description of results from Phase 1 of the Mortality Review
ONS Mortality Seminar (2016). (Office for National Statistics)
Authors: Forster J,
A Comprehensive Framework for Mortality Forecasting
Conference of European Statisticians: Joint Eurostat/UNECE Work Session on Demographic Projections (2016). (Geneva)
Authors: Forster J,
Model integration for mortality estimation and forecasting
Centre for Research in Statistical Methodology Seminar Series (2016). (University of Warwick)
Authors: Forster J,
Stochastic Modelling of UK Mortality Improvement Allowing for Overdispersion
3rd European Actuarial Journal (EAJ) Conference (2016). (Lyon, France)
Authors: Dodd E,
Stochastic Mortality Forecasting with Smoothing and Overdispersion
International Mortality and Longevity Symposium (2016). (Royal Holloway, London)
Authors: Forster J,
English Life Tables No. 17 Methodology
ONS (2015).
Authors: Bijak J, Dodd E, Forster J, Smith P,
Smoothing Mortality Data by Bayesian Model Averaging; Producing the English Life Table
Society of Actuaries 50th Actuarial Research Conference (2015). (Toronto)
Authors: Forster J, Dodd E, Smith P, Bijak J,
Practical Bayesian model averaging: Producing the English Life Table
Seminar (2015). (University of Amsterdam)
Authors: Forster J,