A review and comparison of age-period-cohort models for cancer incidence

Theresa Smith, Jon Wakefield

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Age-period-cohort models have a long and interesting history and area popular choice in a number of disciplines, includingepidemiology, demography, and the social sciences. However, the fitting and interpretation ofthese models requires great care because of the well-knownidentifiability problem that exists; given any two of age, period, andcohort, the third is determined.In this paper we review the identifiability problem and models thathave been proposed for analysis, from both frequentist and Bayesian standpoints. Anumber of recent analyses that use age-period-cohort models aredescribed and critiqued before data on cancer incidence in WashingtonState are analyzed with various models, including a new hybridBayesian approach based on an identifiable parameterization.
Original languageEnglish
Pages (from-to)591-610
JournalStatistical Science
Volume31
Issue number4
DOIs
Publication statusPublished - 19 Jan 2017

Fingerprint

Age-period-cohort Models
Incidence
Cancer
Demography
Identifiability
Social Sciences
Parameterization
Model
Review
Cohort

Keywords

  • age-period-cohort models
  • identifiability
  • random walk priors

Cite this

A review and comparison of age-period-cohort models for cancer incidence. / Smith, Theresa; Wakefield, Jon.

In: Statistical Science, Vol. 31, No. 4, 19.01.2017, p. 591-610.

Research output: Contribution to journalArticle

@article{dd9663fa57d64b62b8c87b177b0bf157,
title = "A review and comparison of age-period-cohort models for cancer incidence",
abstract = "Age-period-cohort models have a long and interesting history and area popular choice in a number of disciplines, includingepidemiology, demography, and the social sciences. However, the fitting and interpretation ofthese models requires great care because of the well-knownidentifiability problem that exists; given any two of age, period, andcohort, the third is determined.In this paper we review the identifiability problem and models thathave been proposed for analysis, from both frequentist and Bayesian standpoints. Anumber of recent analyses that use age-period-cohort models aredescribed and critiqued before data on cancer incidence in WashingtonState are analyzed with various models, including a new hybridBayesian approach based on an identifiable parameterization.",
keywords = "age-period-cohort models, identifiability, random walk priors",
author = "Theresa Smith and Jon Wakefield",
year = "2017",
month = "1",
day = "19",
doi = "10.1214/16-STS580",
language = "English",
volume = "31",
pages = "591--610",
journal = "Statistical Science",
issn = "0883-4237",
publisher = "Institute of Mathematical Statistics",
number = "4",

}

TY - JOUR

T1 - A review and comparison of age-period-cohort models for cancer incidence

AU - Smith, Theresa

AU - Wakefield, Jon

PY - 2017/1/19

Y1 - 2017/1/19

N2 - Age-period-cohort models have a long and interesting history and area popular choice in a number of disciplines, includingepidemiology, demography, and the social sciences. However, the fitting and interpretation ofthese models requires great care because of the well-knownidentifiability problem that exists; given any two of age, period, andcohort, the third is determined.In this paper we review the identifiability problem and models thathave been proposed for analysis, from both frequentist and Bayesian standpoints. Anumber of recent analyses that use age-period-cohort models aredescribed and critiqued before data on cancer incidence in WashingtonState are analyzed with various models, including a new hybridBayesian approach based on an identifiable parameterization.

AB - Age-period-cohort models have a long and interesting history and area popular choice in a number of disciplines, includingepidemiology, demography, and the social sciences. However, the fitting and interpretation ofthese models requires great care because of the well-knownidentifiability problem that exists; given any two of age, period, andcohort, the third is determined.In this paper we review the identifiability problem and models thathave been proposed for analysis, from both frequentist and Bayesian standpoints. Anumber of recent analyses that use age-period-cohort models aredescribed and critiqued before data on cancer incidence in WashingtonState are analyzed with various models, including a new hybridBayesian approach based on an identifiable parameterization.

KW - age-period-cohort models

KW - identifiability

KW - random walk priors

UR - http://projecteuclid.org/euclid.ss

UR - http://dx.doi.org/10.1214/16-STS580

U2 - 10.1214/16-STS580

DO - 10.1214/16-STS580

M3 - Article

VL - 31

SP - 591

EP - 610

JO - Statistical Science

JF - Statistical Science

SN - 0883-4237

IS - 4

ER -