Statistics New Zealand Working Paper No 16–04
John Bryant, Kim Dunstan, Patrick Graham, Nathaniel Matheson-Dunning, Emily Shrosbree, and Richard Speirs
The census-year estimated resident population (ERP) is one of the foundations of New Zealand’s statistical system. Traditionally, national statistical offices have only produced point estimates of the resident population counts. Establishing the level of uncertainty in the ERP is challenging, because computation of the ERP combines information from multiple sources that are each subject to limitations such as missing data and under-coverage. We adopt a Monte Carlo simulation approach to assessing uncertainty. This involves modelling the most-important components of uncertainty and embedding the ERP calculations in a Monte Carlo simulation in which the ERP computation is repeated 1,000 times. The distribution of counts over the 1,000 simulations of the ERP quantifies uncertainty. Particular attention is paid to the modelling of census under-coverage, which is the largest source of uncertainty. At the national level the relative uncertainty in the ERP is less than 1 percent, except at the oldest ages. On the other hand, relative uncertainty in the ERP exceeds 5 percent at some age groups for Māori in smaller regional council areas. These measures of uncertainty provide useful context for customers and may inform future investigations of census coverage survey designs, and research on the potential for administrative data to contribute to population estimation systems.
Key words: uncertainty, population, census coverage, simulation
To read the paper, download or print the PDF from the 'Available files' box. If you have problems viewing the file, see Opening files and PDFs.
Bryant, J, Dunstan, K, Graham, P, Matheson-Dunning, N, Shrosbree, E, & Speirs, R (2016). Measuring uncertainty in the 2013-base estimated resident population (Statistics New Zealand Working Paper No 16-04). Retrieved from www.stats.govt.nz.
ISBN 978-0-908350-63-6 (online)
ISSN 1179-934X (online)
Published 22 September 2016