Identifiability issues of age-period and age-period-cohort models of the Lee-Carter type

DOI10.1016/J.INSMATHECO.2017.04.006zbMATH Open1394.91188arXiv1612.04091OpenAlexW2561082488MaRDI QIDQ2364014FDOQ2364014

Jean-Pierre Urbain, Eric Beutner, Simon Reese

Publication date: 17 July 2017

Published in: Insurance Mathematics \& Economics (Search for Journal in Brave)

Abstract: The predominant way of modelling mortality rates is the Lee-Carter model and its many extensions. The Lee-Carter model and its many extensions use a latent process to forecast. These models are estimated using a two-step procedure that causes an inconsistent view on the latent variable. This paper considers identifiability issues of these models from a perspective that acknowledges the latent variable as a stochastic process from the beginning. We call this perspective the plug-in age-period or plug-in age-period-cohort model. Defining a parameter vector that includes the underlying parameters of this process rather than its realisations, we investigate whether the expected values and covariances of the plug-in Lee-Carter models are identifiable. It will be seen, for example, that even if in both steps of the estimation procedure we have identifiability in a certain sense it does not necessarily carry over to the plug-in models.

Full work available at URL: https://arxiv.org/abs/1612.04091

Recommendations

• Mortality, longevity and experiments with the Lee-Carter model
• On age-period-cohort parametric mortality rate projections
• Inference pitfalls in Lee-Carter model for forecasting mortality
• Age-coherent extensions of the Lee-Carter model
• A cohort-based extension to the Lee-Carter model for mortality reduction factors

zbMATH Keywords

Lee-Carter modelidentifiabilityage-period-cohort modeltime series modelage-period modelplug-in Lee-Carter model

Mathematics Subject Classification ID

Mathematical geography and demography (91D20) Applications of statistics to actuarial sciences and financial mathematics (62P05)

Cites Work

• Identification of the age-period-cohort model and the extended chain-ladder model
• Modeling and Forecasting U.S. Mortality
• Title not available (Why is that?)
• Forecasting with the age-period-cohort model and the extended chain-ladder model
• Title not available (Why is that?)
• Modelling and management of mortality risk: a review
• A Quantitative Comparison of Stochastic Mortality Models Using Data From England and Wales and the United States
• A cohort-based extension to the Lee-Carter model for mortality reduction factors
• Bootstrapping the Poisson log-bilinear model for mortality forecasting
• Title not available (Why is that?)
• On age-period-cohort parametric mortality rate projections
• Extending Lee–Carter Mortality Forecasting
• Identification of causal factor models of stationary time series
• On fitting generalized linear and non-linear models of mortality
• Robustness and convergence in the Lee-Carter model with cohort effects
• Identification and forecasting in mortality models
• Cohort extensions of the Poisson common factor model for modelling both genders jointly
• Mortality, longevity and experiments with the Lee-Carter model
• Inference pitfalls in Lee-Carter model for forecasting mortality

Cited In (8)

• A class of estimable contrasts in an age-period-cohort model
• Longevity risk and capital markets: the 2019--20 update
• A more meaningful parameterization of the Lee-Carter model
• Pitfalls and merits of cointegration-based mortality models
• Identification of the age-period-cohort model and the extended chain-ladder model
• Robustness and convergence in the Lee-Carter model with cohort effects
• Generalised additive dependency inflated models including aggregated covariates
• Title not available (Why is that?)

This page was built for publication: Identifiability issues of age-period and age-period-cohort models of the Lee-Carter type

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2364014)