Predictive Modeling of Threshold Life Tables

Min Ji, Mostafa Aminzadeh and Min Deng

North American Actuarial Journal, 2020, vol. 24, issue 2, 316-332

Abstract: Accurate modeling of mortality at advanced ages is crucial to the valuation of pension plans and life annuities, especially longevity annuities for which annuity payments start very late in life, say age 85. Parametric regression models extrapolate lifetime distribution to the advanced ages but do not guarantee goodness fit. Threshold life tables offer an alternative solution to this problem using piecewise distribution via the peaks-over-threshold approach in the extreme value family. However, parameter estimation of this model is challenging, especially the determination of the threshold. Regular estimation methods do not guarantee a smooth life table. In this research, we impose parameter constraints to achieve a smooth threshold life table and propose a Bayesian approach to obtain Bayes estimates of the parameters and approximate predictive density via simulation, which can be used to compute life expectancy and other measures of interest in a Bayesian framework.

Date: 2020
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/10920277.2019.1633929 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:uaajxx:v:24:y:2020:i:2:p:316-332

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/uaaj20

DOI: 10.1080/10920277.2019.1633929

Access Statistics for this article

North American Actuarial Journal is currently edited by Kathryn Baker

More articles in North American Actuarial Journal from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().