The University of Southampton
University of Southampton Institutional Repository

Forecasting of cohort fertility under a hierarchical Bayesian approach

Forecasting of cohort fertility under a hierarchical Bayesian approach
Forecasting of cohort fertility under a hierarchical Bayesian approach
The main reason for modelling fertility rates is to generate population estimates and forecasts. Previously deterministic in nature, the current fertility literature is favouring stochastic approaches, making models within a Bayesian framework very appealing. In keeping with the recent literature, our aim is to develop a hierarchical Bayesian model for modelling and forecasting cohort fertility rates. Building on the work of Schmertmann et al. (Journal of the American Statistical Association, 2014, pages 500-513), we propose a hierarchical Bayesian approach, using state-of-the-art Hamiltonian Monte Carlo methods and a dataset taken largely from the Human Fertility Database. Obtaining fertility forecasts for 41 countries using the data available in 2010 and 33 countries using the data available in 2000, we use scoring rules to quantitatively assess the predictive accuracy of the latter. The forecasts generated from our model are plausible with appropriate levels of uncertainty, particularly for countries with stable fertility histories in recent years. We illustrate the benefits of taking a hierarchical Bayesian approach to modelling cohort fertility. The flexibility of the approach allows us to incorporate the natural structure of fertility data indexed by age and cohort, and the underlying assumptions in both dimensions, intuitively and transparently. In addition, it demonstrates how advanced computational methods can be used to fit hierarchical Bayesian models with an atypical setup. This not only cements the position of hierarchical Bayesian methods at the forefront of population forecasting methods, but also makes a valuable contribution to the fertility modelling and forecasting literature.
University of Southampton
Ellison, Joanne
c81d4cd0-3c0a-4a3a-878b-206fb27b7b31
Forster, Jonathan
e3c534ad-fa69-42f5-b67b-11617bc84879
Dodd, Erengul
b3faed76-f22b-4928-a922-7f0b8439030d
Ellison, Joanne
c81d4cd0-3c0a-4a3a-878b-206fb27b7b31
Forster, Jonathan
e3c534ad-fa69-42f5-b67b-11617bc84879
Dodd, Erengul
b3faed76-f22b-4928-a922-7f0b8439030d

Ellison, Joanne, Forster, Jonathan and Dodd, Erengul (2018) Forecasting of cohort fertility under a hierarchical Bayesian approach University of Southampton

Record type: Monograph (Working Paper)

Abstract

The main reason for modelling fertility rates is to generate population estimates and forecasts. Previously deterministic in nature, the current fertility literature is favouring stochastic approaches, making models within a Bayesian framework very appealing. In keeping with the recent literature, our aim is to develop a hierarchical Bayesian model for modelling and forecasting cohort fertility rates. Building on the work of Schmertmann et al. (Journal of the American Statistical Association, 2014, pages 500-513), we propose a hierarchical Bayesian approach, using state-of-the-art Hamiltonian Monte Carlo methods and a dataset taken largely from the Human Fertility Database. Obtaining fertility forecasts for 41 countries using the data available in 2010 and 33 countries using the data available in 2000, we use scoring rules to quantitatively assess the predictive accuracy of the latter. The forecasts generated from our model are plausible with appropriate levels of uncertainty, particularly for countries with stable fertility histories in recent years. We illustrate the benefits of taking a hierarchical Bayesian approach to modelling cohort fertility. The flexibility of the approach allows us to incorporate the natural structure of fertility data indexed by age and cohort, and the underlying assumptions in both dimensions, intuitively and transparently. In addition, it demonstrates how advanced computational methods can be used to fit hierarchical Bayesian models with an atypical setup. This not only cements the position of hierarchical Bayesian methods at the forefront of population forecasting methods, but also makes a valuable contribution to the fertility modelling and forecasting literature.

Text
ELLISON_paper - Author's Original
Download (1MB)

More information

Published date: 2018

Identifiers

Local EPrints ID: 424927
URI: https://eprints.soton.ac.uk/id/eprint/424927
PURE UUID: dead073b-5d4c-47af-b4bc-13c9dfdea9e6

Catalogue record

Date deposited: 05 Oct 2018 16:30
Last modified: 13 Mar 2019 17:59

Export record

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×