Block Constraints in Age–Period–Cohort Models with Unequal-width Intervals

Liying Luo, James S. Hodges

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Age–period–cohort (APC) models are designed to estimate the independent effects of age, time periods, and cohort membership. However, APC models suffer from an identification problem: There are no unique estimates of the independent effects that fit the data best because of the exact linear dependency among age, period, and cohort. Among methods proposed to address this problem, using unequal-interval widths for age, period, and cohort categories appears to break the exact linear dependency and thus solve the identification problem. However, this article shows that the identification problem remains in these models; in fact, they just implicitly impose multiple block constraints on the age, period, and cohort effects to achieve identifiability. These constraints depend on an arbitrary choice of widths for the age, period, and cohort intervals and can have nontrivial effects on the estimates. Because these assumptions are extremely difficult, if not impossible, to verify in empirical research, they are qualitatively no different from the assumptions of other constrained estimators. Therefore, the unequal-intervals approach should not be used without an explicit rationale justifying their constraints.

Original languageEnglish (US)
Pages (from-to)700-726
Number of pages27
JournalSociological Methods and Research
Volume45
Issue number4
DOIs
StatePublished - Nov 1 2016

Fingerprint

empirical research
time

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)
  • Sociology and Political Science

Cite this

@article{9084e0228578473586884e50133f2d3b,
title = "Block Constraints in Age–Period–Cohort Models with Unequal-width Intervals",
abstract = "Age–period–cohort (APC) models are designed to estimate the independent effects of age, time periods, and cohort membership. However, APC models suffer from an identification problem: There are no unique estimates of the independent effects that fit the data best because of the exact linear dependency among age, period, and cohort. Among methods proposed to address this problem, using unequal-interval widths for age, period, and cohort categories appears to break the exact linear dependency and thus solve the identification problem. However, this article shows that the identification problem remains in these models; in fact, they just implicitly impose multiple block constraints on the age, period, and cohort effects to achieve identifiability. These constraints depend on an arbitrary choice of widths for the age, period, and cohort intervals and can have nontrivial effects on the estimates. Because these assumptions are extremely difficult, if not impossible, to verify in empirical research, they are qualitatively no different from the assumptions of other constrained estimators. Therefore, the unequal-intervals approach should not be used without an explicit rationale justifying their constraints.",
author = "Liying Luo and Hodges, {James S.}",
year = "2016",
month = "11",
day = "1",
doi = "10.1177/0049124115585359",
language = "English (US)",
volume = "45",
pages = "700--726",
journal = "Sociological Methods and Research",
issn = "0049-1241",
publisher = "SAGE Publications Inc.",
number = "4",

}

Block Constraints in Age–Period–Cohort Models with Unequal-width Intervals. / Luo, Liying; Hodges, James S.

In: Sociological Methods and Research, Vol. 45, No. 4, 01.11.2016, p. 700-726.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Block Constraints in Age–Period–Cohort Models with Unequal-width Intervals

AU - Luo, Liying

AU - Hodges, James S.

PY - 2016/11/1

Y1 - 2016/11/1

N2 - Age–period–cohort (APC) models are designed to estimate the independent effects of age, time periods, and cohort membership. However, APC models suffer from an identification problem: There are no unique estimates of the independent effects that fit the data best because of the exact linear dependency among age, period, and cohort. Among methods proposed to address this problem, using unequal-interval widths for age, period, and cohort categories appears to break the exact linear dependency and thus solve the identification problem. However, this article shows that the identification problem remains in these models; in fact, they just implicitly impose multiple block constraints on the age, period, and cohort effects to achieve identifiability. These constraints depend on an arbitrary choice of widths for the age, period, and cohort intervals and can have nontrivial effects on the estimates. Because these assumptions are extremely difficult, if not impossible, to verify in empirical research, they are qualitatively no different from the assumptions of other constrained estimators. Therefore, the unequal-intervals approach should not be used without an explicit rationale justifying their constraints.

AB - Age–period–cohort (APC) models are designed to estimate the independent effects of age, time periods, and cohort membership. However, APC models suffer from an identification problem: There are no unique estimates of the independent effects that fit the data best because of the exact linear dependency among age, period, and cohort. Among methods proposed to address this problem, using unequal-interval widths for age, period, and cohort categories appears to break the exact linear dependency and thus solve the identification problem. However, this article shows that the identification problem remains in these models; in fact, they just implicitly impose multiple block constraints on the age, period, and cohort effects to achieve identifiability. These constraints depend on an arbitrary choice of widths for the age, period, and cohort intervals and can have nontrivial effects on the estimates. Because these assumptions are extremely difficult, if not impossible, to verify in empirical research, they are qualitatively no different from the assumptions of other constrained estimators. Therefore, the unequal-intervals approach should not be used without an explicit rationale justifying their constraints.

UR - http://www.scopus.com/inward/record.url?scp=84990181629&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84990181629&partnerID=8YFLogxK

U2 - 10.1177/0049124115585359

DO - 10.1177/0049124115585359

M3 - Article

AN - SCOPUS:84990181629

VL - 45

SP - 700

EP - 726

JO - Sociological Methods and Research

JF - Sociological Methods and Research

SN - 0049-1241

IS - 4

ER -