Statistics New Zealand prepares regular population projections at the subnational level. Like other statistical agencies, Statistics NZ currently uses a deterministic approach to subnational projections. However, we are currently developing an alternative, more statistical, approach. In this paper, we describe two Bayesian hierarchical models for estimating and forecasting international in-migration and out-migration rates, disaggregated by age, sex, and territorial authority. The model uses international arrivals and departures data for 1991-2013, and produces forecasts for 2014-2039. Special attention is given to the ability of the model to estimate migration rates for small population domains. The modelled approach provides a coherent and integrated measure of uncertainty at the detailed demographic level. Examples of estimated and forecasted migration rates are presented. We conclude with a discussion of the challenges and limitations of the data and model, and directions for future developments.
[1]
Jakub Bijak,et al.
Bayesian forecasting of immigration to selected European countries by using expert knowledge
,
2010
.
[2]
Radford M. Neal.
MCMC Using Hamiltonian Dynamics
,
2011,
1206.1901.
[3]
A. Gelman.
Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper)
,
2004
.
[4]
J. Poot,et al.
A Stochastic Sub-national Population Projection Methodology with an Application to the Waikato Region of New Zealand
,
2010
.
[5]
Adrian E. Raftery,et al.
Bayesian Probabilistic Projection of International Migration Rates
,
2013,
1310.7148.
[6]
D. Rubin.
Multiple imputation for nonresponse in surveys
,
1989
.