Gamma-Gompertz life expectancy at birth
BACKGROUND The gamma-Gompertz multiplicative frailty model is the most common parametric modelapplied to human mortality data at adult and old ages. The resulting life expectancy hasbeen calculated so far only numerically. OBJECTIVE Properties of the gamma-Gompertz distribution have not been thoroug...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
Max Planck Institute for Demographic Research
2013-02-01
|
Series: | Demographic Research |
Subjects: | |
Online Access: | http://www.demographic-research.org/volumes/vol28/9/ |
_version_ | 1818084486603079680 |
---|---|
author | Trifon I. Missov |
author_facet | Trifon I. Missov |
author_sort | Trifon I. Missov |
collection | DOAJ |
description | BACKGROUND The gamma-Gompertz multiplicative frailty model is the most common parametric modelapplied to human mortality data at adult and old ages. The resulting life expectancy hasbeen calculated so far only numerically. OBJECTIVE Properties of the gamma-Gompertz distribution have not been thoroughly studied. The focusof the paper is to shed light onto its first moment or, demographically speaking, characterizelife expectancy resulting from a gamma-Gompertz force of mortality. The paperprovides an exact formula for gamma-Gompertz life expectancy at birth and a simplerhigh-accuracy approximation that can be used in practice for computational convenience.In addition, the article compares actual (life-table) to model-based (gamma-Gompertz)life expectancy to assess on aggregate how many years of life expectancy are not captured(or overestimated) by the gamma-Gompertz mortality mechanism. COMMENTS A closed-form expression for gamma-Gomeprtz life expectancy at birth contains a special(the hypergeometric) function. It aids assessing the impact of gamma-Gompertz parameterson life expectancy values. The paper shows that a high-accuracy approximation canbe constructed by assuming an integer value for the shape parameter of the gamma distribution.A historical comparison between model-based and actual life expectancy forSwedish females reveals a gap that is decreasing to around 2 years from 1950 onwards.Looking at remaining life expectancies at ages 30 and 50, we see this gap almost disappearing. |
first_indexed | 2024-12-10T19:54:39Z |
format | Article |
id | doaj.art-f79427987bf44f3a96c9a1607155275b |
institution | Directory Open Access Journal |
issn | 1435-9871 |
language | English |
last_indexed | 2024-12-10T19:54:39Z |
publishDate | 2013-02-01 |
publisher | Max Planck Institute for Demographic Research |
record_format | Article |
series | Demographic Research |
spelling | doaj.art-f79427987bf44f3a96c9a1607155275b2022-12-22T01:35:41ZengMax Planck Institute for Demographic ResearchDemographic Research1435-98712013-02-01289Gamma-Gompertz life expectancy at birthTrifon I. MissovBACKGROUND The gamma-Gompertz multiplicative frailty model is the most common parametric modelapplied to human mortality data at adult and old ages. The resulting life expectancy hasbeen calculated so far only numerically. OBJECTIVE Properties of the gamma-Gompertz distribution have not been thoroughly studied. The focusof the paper is to shed light onto its first moment or, demographically speaking, characterizelife expectancy resulting from a gamma-Gompertz force of mortality. The paperprovides an exact formula for gamma-Gompertz life expectancy at birth and a simplerhigh-accuracy approximation that can be used in practice for computational convenience.In addition, the article compares actual (life-table) to model-based (gamma-Gompertz)life expectancy to assess on aggregate how many years of life expectancy are not captured(or overestimated) by the gamma-Gompertz mortality mechanism. COMMENTS A closed-form expression for gamma-Gomeprtz life expectancy at birth contains a special(the hypergeometric) function. It aids assessing the impact of gamma-Gompertz parameterson life expectancy values. The paper shows that a high-accuracy approximation canbe constructed by assuming an integer value for the shape parameter of the gamma distribution.A historical comparison between model-based and actual life expectancy forSwedish females reveals a gap that is decreasing to around 2 years from 1950 onwards.Looking at remaining life expectancies at ages 30 and 50, we see this gap almost disappearing.http://www.demographic-research.org/volumes/vol28/9/approximationsgamma-Gompertz frailty modelhypergeometric functionlife expectancy at birth |
spellingShingle | Trifon I. Missov Gamma-Gompertz life expectancy at birth Demographic Research approximations gamma-Gompertz frailty model hypergeometric function life expectancy at birth |
title | Gamma-Gompertz life expectancy at birth |
title_full | Gamma-Gompertz life expectancy at birth |
title_fullStr | Gamma-Gompertz life expectancy at birth |
title_full_unstemmed | Gamma-Gompertz life expectancy at birth |
title_short | Gamma-Gompertz life expectancy at birth |
title_sort | gamma gompertz life expectancy at birth |
topic | approximations gamma-Gompertz frailty model hypergeometric function life expectancy at birth |
url | http://www.demographic-research.org/volumes/vol28/9/ |
work_keys_str_mv | AT trifonimissov gammagompertzlifeexpectancyatbirth |