A functional data analysis approach for forecasting population: a case study for the United Kingdom

Abstract

Cohort component models are often used to model the evolution of an age-specific population, and are particularly useful to highlight which demographic component contributes the most to population change. Many methods have been proposed to forecast four demographic components, namely mortality, fertility, emigration and immigration. These existing methods are sometimes considered from a deterministic viewpoint, which in practice can be quite restrictive. The statistical method we propose is a multilevel functional data analytic approach, where the mortality and migration for females and males are modelled and forecasted jointly. The forecast uncertainty associated with each component is incorporated through bootstrapping. Using the historical data for the United Kingdom from 1975 to 2009, we found that the proposed method shows good in-sample forecast accuracy for the holdout data between years 2000 and 2009. Moreover, we produce out-of-sample population forecasts from 2010 to 2030, and compare our forecasts with those produced by the Office for National Statistics

More information

Identifiers

ORCID for Peter W.F. Smith: orcid.org/0000-0003-4423-5410

ORCID for Jakub Bijak: orcid.org/0000-0002-2563-5040

Catalogue record

Export record