Coherent Modeling and Forecasting of Mortality Patterns for Subpopulations Using Multiway Analysis of Compositions

An Application to Canadian Provinces and Territories

Marie-Pier Bergeron Boucher*, Violetta Simonacci, James Oeppen, Michele Gallo

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Mortality levels for subpopulations, such as countries in a region or provinces within a country, generally change in a similar fashion over time, as a result of common historical experiences in terms of health, culture, and economics. Forecasting mortality for such populations should consider the correlation between their mortality levels. In this perspective, we suggest using multilinear component techniques to identify a common time trend and then use it to forecast coherently the mortality of subpopulations. Moreover, this multiway approach is performed on life table deaths by referring to Compositional Data Analysis (CoDa) methodology. Compositional data are strictly positive values summing to a constant and represent part of a whole. Life table deaths are compositional by definition because they provide the age composition of deaths per year and sum to the life table radix. In bilinear models the use of life table deaths treated as compositions generally leads to less biased forecasts than other commonly used models by not assuming a constant rate of mortality improvement. As a consequence, an extension of this approach to multiway data is here presented. Specifically, a CoDa adaptation of the Tucker3 model is implemented for life table deaths arranged in three-dimensional arrays indexed by time, age, and population. The proposed procedure is used to forecast the mortality of Canadian provinces in a comparative study. The results show that the proposed model leads to coherent forecasts.

Original languageEnglish
JournalThe North American Actuarial Journal
Volume22
Issue number1
Pages (from-to)92-118
ISSN1092-0277
DOIs
Publication statusPublished - 10. Jan 2018

Fingerprint

Multiway Analysis
Life Table
Mortality
Forecasting
Compositional Data
Forecast
Modeling
Data analysis
Tucker3
Bilinear Model
Strictly positive
Rate Constant
Comparative Study
Biased
Life table
Health
Model
Economics
Three-dimensional
Methodology

Cite this

@article{b19b12e09b8b406e92566cb34d8ef75b,
title = "Coherent Modeling and Forecasting of Mortality Patterns for Subpopulations Using Multiway Analysis of Compositions: An Application to Canadian Provinces and Territories",
abstract = "Mortality levels for subpopulations, such as countries in a region or provinces within a country, generally change in a similar fashion over time, as a result of common historical experiences in terms of health, culture, and economics. Forecasting mortality for such populations should consider the correlation between their mortality levels. In this perspective, we suggest using multilinear component techniques to identify a common time trend and then use it to forecast coherently the mortality of subpopulations. Moreover, this multiway approach is performed on life table deaths by referring to Compositional Data Analysis (CoDa) methodology. Compositional data are strictly positive values summing to a constant and represent part of a whole. Life table deaths are compositional by definition because they provide the age composition of deaths per year and sum to the life table radix. In bilinear models the use of life table deaths treated as compositions generally leads to less biased forecasts than other commonly used models by not assuming a constant rate of mortality improvement. As a consequence, an extension of this approach to multiway data is here presented. Specifically, a CoDa adaptation of the Tucker3 model is implemented for life table deaths arranged in three-dimensional arrays indexed by time, age, and population. The proposed procedure is used to forecast the mortality of Canadian provinces in a comparative study. The results show that the proposed model leads to coherent forecasts.",
author = "{Bergeron Boucher}, Marie-Pier and Violetta Simonacci and James Oeppen and Michele Gallo",
year = "2018",
month = "1",
day = "10",
doi = "10.1080/10920277.2017.1377620",
language = "English",
volume = "22",
pages = "92--118",
journal = "The North American Actuarial Journal",
issn = "1092-0277",
publisher = "Heinemann",
number = "1",

}

Coherent Modeling and Forecasting of Mortality Patterns for Subpopulations Using Multiway Analysis of Compositions : An Application to Canadian Provinces and Territories. / Bergeron Boucher, Marie-Pier; Simonacci, Violetta; Oeppen, James; Gallo, Michele.

In: The North American Actuarial Journal, Vol. 22, No. 1, 10.01.2018, p. 92-118.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Coherent Modeling and Forecasting of Mortality Patterns for Subpopulations Using Multiway Analysis of Compositions

T2 - An Application to Canadian Provinces and Territories

AU - Bergeron Boucher, Marie-Pier

AU - Simonacci, Violetta

AU - Oeppen, James

AU - Gallo, Michele

PY - 2018/1/10

Y1 - 2018/1/10

N2 - Mortality levels for subpopulations, such as countries in a region or provinces within a country, generally change in a similar fashion over time, as a result of common historical experiences in terms of health, culture, and economics. Forecasting mortality for such populations should consider the correlation between their mortality levels. In this perspective, we suggest using multilinear component techniques to identify a common time trend and then use it to forecast coherently the mortality of subpopulations. Moreover, this multiway approach is performed on life table deaths by referring to Compositional Data Analysis (CoDa) methodology. Compositional data are strictly positive values summing to a constant and represent part of a whole. Life table deaths are compositional by definition because they provide the age composition of deaths per year and sum to the life table radix. In bilinear models the use of life table deaths treated as compositions generally leads to less biased forecasts than other commonly used models by not assuming a constant rate of mortality improvement. As a consequence, an extension of this approach to multiway data is here presented. Specifically, a CoDa adaptation of the Tucker3 model is implemented for life table deaths arranged in three-dimensional arrays indexed by time, age, and population. The proposed procedure is used to forecast the mortality of Canadian provinces in a comparative study. The results show that the proposed model leads to coherent forecasts.

AB - Mortality levels for subpopulations, such as countries in a region or provinces within a country, generally change in a similar fashion over time, as a result of common historical experiences in terms of health, culture, and economics. Forecasting mortality for such populations should consider the correlation between their mortality levels. In this perspective, we suggest using multilinear component techniques to identify a common time trend and then use it to forecast coherently the mortality of subpopulations. Moreover, this multiway approach is performed on life table deaths by referring to Compositional Data Analysis (CoDa) methodology. Compositional data are strictly positive values summing to a constant and represent part of a whole. Life table deaths are compositional by definition because they provide the age composition of deaths per year and sum to the life table radix. In bilinear models the use of life table deaths treated as compositions generally leads to less biased forecasts than other commonly used models by not assuming a constant rate of mortality improvement. As a consequence, an extension of this approach to multiway data is here presented. Specifically, a CoDa adaptation of the Tucker3 model is implemented for life table deaths arranged in three-dimensional arrays indexed by time, age, and population. The proposed procedure is used to forecast the mortality of Canadian provinces in a comparative study. The results show that the proposed model leads to coherent forecasts.

U2 - 10.1080/10920277.2017.1377620

DO - 10.1080/10920277.2017.1377620

M3 - Journal article

VL - 22

SP - 92

EP - 118

JO - The North American Actuarial Journal

JF - The North American Actuarial Journal

SN - 1092-0277

IS - 1

ER -