A Quantitative Comparison of Stochastic Mortality Models Using Data From England and Wales and the United States

By Andrew J. G. Cairns, David P. Blake, Kevin Dowd, Guy D. Coughlan, D. Epstein, Alen Ong, Igor Balevich. 2009

714 citations recorded by the source corpus.

Abstract We compare quantitatively eight stochastic models explaining improvements in mortality rates in England and Wales and in the United States. On the basis of the Bayes Information Criterion (BIC), we find that, for higher ages, an extension of the Cairns-Blake-Dowd (CBD) model that incorporates a cohort effect fits the England and Wales males data best, while for U.S. males data, the Renshaw and Haberman (RH) extension to the Lee and Carter model that also allows for a cohort effect provides the best fit. However, we identify problems with the robustness of parameter estimates under the RH model, calling into question its suitability for forecasting. A different extension to the CBD model that allows not only for a cohort effect, but also for a quadratic age effect, while ranking below the other models in terms of the BIC, exhibits parameter stability across different time periods for both datasets. This model also shows, for both datasets, that there have been approximately linear improvements over time in mortality rates at all ages, but that the improvements have been greater at lower ages than at higher ages, and that there are significant cohort effects.

Subfields: Demography, General Health Professions, Health

View Paper: https://www.semanticscholar.org/paper/a88c648be0197425192704ab6775306039644b94

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Demography

In General Health Professions

In Health

Citations in the corpus

Forecasting: theory and practice (2020, methodology): “., 2011). Among the parametric methods, the method of Heligman and Pollard (1980) is well-known. Among the nonparametric methods, the Lee-Carter model (Lee and Carter, 1992), Cairns-Blake-Dowd model (Cairns et al., 2009; Dowd et al., 2010), and functional data model (Hyndman and Ullah, 2007), as wel”
Measuring Basis Risk in Longevity Hedges (2011, methodology): “This selection criterion is also used by Cairns et al. (2009) in a quantitative comparison of stochastic models.”
Machine learning techniques for mortality modeling (2017, background): “For a broad overview and comparison of existing models we refer to [3].”
Pricing Standardized Mortality Securitizations: A Two‐Population Model with Transitory Jump Effects (2013, neutral): “They are used in the original paper of Lee and Carter (1992) as well as numerous other papers on stochastic mortality modeling (e.g. Cairns et al., 2009).”
A step-by-step guide to building two-population stochastic mortality models (2015, methodology): “In an analysis based on data from the US and EW populations, Cairns et al. (2009) found that Model M7 (the Cairns-Blake-Dowd model with cohort and quadratic effects) performs better than the LC and APC structures in terms of the Bayesian Information Criterion and the robustness of parameter…”
Understanding, modelling and managing longevity risk: key issues and main challenges (2012, background): “In recent contemporary studies, many authors have introduced stochastic models to capture the cohort effect (see e.g. Cairns et al. (2006, 2009)); this subsection, briefly presents some of them.”
Modeling Period Effects in Multi-Population Mortality Models: Applications to Solvency II (2014, neutral): “Depending on the data and evaluation criteria, some more sophisticated models may outperform it (see, e.g., Cairns et al. 2009).”
The New Life Market (2013, methodology): “47 The CBD model with a cohort effect is also known as Model M7, a naming convention introduced in Cairns et al. (2009).”
Modelling and management of longevity risk: Approximations to survivor functions and dynamic hedging (2011, methodology): “…this can be avoided through the use of market models (see, for example, Cairns et al., 2006b, 2008, Cairns, 2007, and Zhu and Bauer, 2010) but such models are less realistic, at present, than the more popular class of “short-rate” models such as those considered in Cairns et al. (2009).”
A comparative study of parametric mortality projection models (2011, methodology): “The age restriction is imposed in order to accommodate the models due to Cairns et al. (2009), denoted by M5-M8, and which are designed for use at pensioner ages only.”