Bivariate smoothing of mortality surfaces with cohort and period ridges

Alexander Dokumentov, Rob J. Hyndman, Leonie Tickle

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)

Abstract

Mortality rates typically vary smoothly over age and time. Exceptions occur owing to events such as wars and epidemics, which create ridges in the age-period surface of mortality rates in a particular year or for cohorts born in a particular year. We propose a new practical method for modelling the age-period surface of mortality rates. Our approach uses L1 regularization with bivariate smoothing and allows for age-varying period and cohort ridges in the otherwise smooth surface. Cross-validation on data from many countries and from simulations demonstrates that our approach is superior to existing approaches in estimating the “true” age-period mortality surface. It also provides greater insight into the underlying mortality dynamics, informing mortality modelling, analysis and forecasting. Although designed for the modelling of mortality rates, our method can also be applied to any bivariate data with occasional ridges and extends the statistical literature on quantile smoothing.

Original languageEnglish
Article numbere199
Number of pages13
JournalStat
Volume7
DOIs
Publication statusPublished - 1 Jan 2018

Keywords

  • bivariate data
  • cohort effects
  • graduation
  • nonparametric smoothing
  • period effects

Cite this

Dokumentov, Alexander ; Hyndman, Rob J. ; Tickle, Leonie. / Bivariate smoothing of mortality surfaces with cohort and period ridges. In: Stat. 2018 ; Vol. 7.
@article{774f37b03b58404880a31bad8e21f705,
title = "Bivariate smoothing of mortality surfaces with cohort and period ridges",
abstract = "Mortality rates typically vary smoothly over age and time. Exceptions occur owing to events such as wars and epidemics, which create ridges in the age-period surface of mortality rates in a particular year or for cohorts born in a particular year. We propose a new practical method for modelling the age-period surface of mortality rates. Our approach uses L1 regularization with bivariate smoothing and allows for age-varying period and cohort ridges in the otherwise smooth surface. Cross-validation on data from many countries and from simulations demonstrates that our approach is superior to existing approaches in estimating the “true” age-period mortality surface. It also provides greater insight into the underlying mortality dynamics, informing mortality modelling, analysis and forecasting. Although designed for the modelling of mortality rates, our method can also be applied to any bivariate data with occasional ridges and extends the statistical literature on quantile smoothing.",
keywords = "bivariate data, cohort effects, graduation, nonparametric smoothing, period effects",
author = "Alexander Dokumentov and Hyndman, {Rob J.} and Leonie Tickle",
year = "2018",
month = "1",
day = "1",
doi = "10.1002/sta4.199",
language = "English",
volume = "7",
journal = "Stat",
issn = "2049-1573",
publisher = "Wiley-Blackwell",

}

Bivariate smoothing of mortality surfaces with cohort and period ridges. / Dokumentov, Alexander; Hyndman, Rob J.; Tickle, Leonie.

In: Stat, Vol. 7, e199, 01.01.2018.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Bivariate smoothing of mortality surfaces with cohort and period ridges

AU - Dokumentov, Alexander

AU - Hyndman, Rob J.

AU - Tickle, Leonie

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Mortality rates typically vary smoothly over age and time. Exceptions occur owing to events such as wars and epidemics, which create ridges in the age-period surface of mortality rates in a particular year or for cohorts born in a particular year. We propose a new practical method for modelling the age-period surface of mortality rates. Our approach uses L1 regularization with bivariate smoothing and allows for age-varying period and cohort ridges in the otherwise smooth surface. Cross-validation on data from many countries and from simulations demonstrates that our approach is superior to existing approaches in estimating the “true” age-period mortality surface. It also provides greater insight into the underlying mortality dynamics, informing mortality modelling, analysis and forecasting. Although designed for the modelling of mortality rates, our method can also be applied to any bivariate data with occasional ridges and extends the statistical literature on quantile smoothing.

AB - Mortality rates typically vary smoothly over age and time. Exceptions occur owing to events such as wars and epidemics, which create ridges in the age-period surface of mortality rates in a particular year or for cohorts born in a particular year. We propose a new practical method for modelling the age-period surface of mortality rates. Our approach uses L1 regularization with bivariate smoothing and allows for age-varying period and cohort ridges in the otherwise smooth surface. Cross-validation on data from many countries and from simulations demonstrates that our approach is superior to existing approaches in estimating the “true” age-period mortality surface. It also provides greater insight into the underlying mortality dynamics, informing mortality modelling, analysis and forecasting. Although designed for the modelling of mortality rates, our method can also be applied to any bivariate data with occasional ridges and extends the statistical literature on quantile smoothing.

KW - bivariate data

KW - cohort effects

KW - graduation

KW - nonparametric smoothing

KW - period effects

UR - http://www.scopus.com/inward/record.url?scp=85052797230&partnerID=8YFLogxK

U2 - 10.1002/sta4.199

DO - 10.1002/sta4.199

M3 - Article

VL - 7

JO - Stat

JF - Stat

SN - 2049-1573

M1 - e199

ER -