A hierarchical model for the joint mortality analysis of pension scheme data with missing covariates

Francesco Ungolo (Corresponding author), Torsten Kleinow, Angus S. Macdonald

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

A hierarchical model is developed for the joint mortality analysis of pension scheme datasets. The proposed model allows for a rigorous statistical treatment of missing data. While our approach works for any missing data pattern, we are particularly interested in a scenario where some covariates are observed for members of one pension scheme but not the other. Therefore, our approach allows for the joint modelling of datasets which contain different information about individual lives. The proposed model generalizes the specification of parametric models when accounting for covariates. We consider parameter uncertainty using Bayesian techniques. Model parametrization is analysed in order to obtain an efficient MCMC sampler, and address model selection. The inferential framework described here accommodates any missing-data pattern, and turns out to be useful to analyse statistical relationships among covariates. Finally, we assess the financial impact of using the covariates, and of the optimal use of the whole available sample when combining data from different mortality experiences.
Original languageEnglish
Pages (from-to)68-84
Number of pages17
JournalInsurance: Mathematics and Economics
Volume91
DOIs
Publication statusPublished - Mar 2020

Fingerprint

Missing Covariates
Hierarchical Model
Mortality
Covariates
Missing Data
Joint Modeling
Parameter Uncertainty
Markov Chain Monte Carlo
Parametric Model
Model Selection
Parametrization
Model
Specification
Scenarios
Generalise
Hierarchical model
Pension scheme
Missing data

Keywords

  • Bayesian inference
  • Longevity risk
  • MCMC
  • Missing data
  • Model selection
  • Mortality
  • Mortality models with covariates
  • Survival model

Cite this

@article{0434163e5b3b49b2af04907688b0a394,
title = "A hierarchical model for the joint mortality analysis of pension scheme data with missing covariates",
abstract = "A hierarchical model is developed for the joint mortality analysis of pension scheme datasets. The proposed model allows for a rigorous statistical treatment of missing data. While our approach works for any missing data pattern, we are particularly interested in a scenario where some covariates are observed for members of one pension scheme but not the other. Therefore, our approach allows for the joint modelling of datasets which contain different information about individual lives. The proposed model generalizes the specification of parametric models when accounting for covariates. We consider parameter uncertainty using Bayesian techniques. Model parametrization is analysed in order to obtain an efficient MCMC sampler, and address model selection. The inferential framework described here accommodates any missing-data pattern, and turns out to be useful to analyse statistical relationships among covariates. Finally, we assess the financial impact of using the covariates, and of the optimal use of the whole available sample when combining data from different mortality experiences.",
keywords = "Bayesian inference, Longevity risk, MCMC, Missing data, Model selection, Mortality, Mortality models with covariates, Survival model",
author = "Francesco Ungolo and Torsten Kleinow and Macdonald, {Angus S.}",
year = "2020",
month = "3",
doi = "10.1016/j.insmatheco.2020.01.003",
language = "English",
volume = "91",
pages = "68--84",
journal = "Insurance: Mathematics and Economics",
issn = "0167-6687",
publisher = "Elsevier",

}

A hierarchical model for the joint mortality analysis of pension scheme data with missing covariates. / Ungolo, Francesco (Corresponding author); Kleinow, Torsten; Macdonald, Angus S.

In: Insurance: Mathematics and Economics, Vol. 91, 03.2020, p. 68-84.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - A hierarchical model for the joint mortality analysis of pension scheme data with missing covariates

AU - Ungolo, Francesco

AU - Kleinow, Torsten

AU - Macdonald, Angus S.

PY - 2020/3

Y1 - 2020/3

N2 - A hierarchical model is developed for the joint mortality analysis of pension scheme datasets. The proposed model allows for a rigorous statistical treatment of missing data. While our approach works for any missing data pattern, we are particularly interested in a scenario where some covariates are observed for members of one pension scheme but not the other. Therefore, our approach allows for the joint modelling of datasets which contain different information about individual lives. The proposed model generalizes the specification of parametric models when accounting for covariates. We consider parameter uncertainty using Bayesian techniques. Model parametrization is analysed in order to obtain an efficient MCMC sampler, and address model selection. The inferential framework described here accommodates any missing-data pattern, and turns out to be useful to analyse statistical relationships among covariates. Finally, we assess the financial impact of using the covariates, and of the optimal use of the whole available sample when combining data from different mortality experiences.

AB - A hierarchical model is developed for the joint mortality analysis of pension scheme datasets. The proposed model allows for a rigorous statistical treatment of missing data. While our approach works for any missing data pattern, we are particularly interested in a scenario where some covariates are observed for members of one pension scheme but not the other. Therefore, our approach allows for the joint modelling of datasets which contain different information about individual lives. The proposed model generalizes the specification of parametric models when accounting for covariates. We consider parameter uncertainty using Bayesian techniques. Model parametrization is analysed in order to obtain an efficient MCMC sampler, and address model selection. The inferential framework described here accommodates any missing-data pattern, and turns out to be useful to analyse statistical relationships among covariates. Finally, we assess the financial impact of using the covariates, and of the optimal use of the whole available sample when combining data from different mortality experiences.

KW - Bayesian inference

KW - Longevity risk

KW - MCMC

KW - Missing data

KW - Model selection

KW - Mortality

KW - Mortality models with covariates

KW - Survival model

UR - http://www.scopus.com/inward/record.url?scp=85077925666&partnerID=8YFLogxK

U2 - 10.1016/j.insmatheco.2020.01.003

DO - 10.1016/j.insmatheco.2020.01.003

M3 - Article

VL - 91

SP - 68

EP - 84

JO - Insurance: Mathematics and Economics

JF - Insurance: Mathematics and Economics

SN - 0167-6687

ER -