Crystal Cy Yu

Population projections are used by a number of local agencies to better prepare for the future resource needs of counties, ensuring that educational, health, housing, and economic demands of individuals are met. Meeting the specific needs of a county’s population, such as what resources to provide, where to target resources, and ensure an equitable distribution of those resources, requires population projections which are both demographically detailed, such as by age, race, and ethnicity, and geographically precise, such as at the census tract level. Despite this need, an evaluation of which methods are best suited to produce population projections at this level are lacking. In this study, we evaluate the accuracy of several cohort-based methods for small area population projections by race and ethnicity. We apply these methods to population projections of King County, Washington and assess the validity of projections using past population estimates. We find a clear pattern that demonstrates while simplified methods perform well in near term forecasts, methods which employ smoothing strategies perform better in long-term forecasting scenarios. Furthermore, we demonstrate that model’s incorporating multiple stages of smoothing can provide detailed insights into the projected population size of King county and the places and groups which will most contribute to this growth. Detailed projections, such as those provided by multi-stage smoothing methods, enable city planners and policy makers a detailed view of the future structure of their county’s population and provide for them a resource to better meet the needs of future populations.

Population projections provide predictions of future population sizes for an area. Historically, most population projections have been produced using deterministic or scenario-based approaches and have not assessed uncertainty about future population change. Starting in 2015, however, the United Nations (UN) has produced probabilistic population projections for all countries using a Bayesian approach. There is also considerable interest in subnational probabilistic population projections, but the UN's national approach cannot be used directly for this purpose, because within-country correlations in fertility and mortality are generally larger than between-country ones, migration is not constrained in the same way, and there is a need to account for college and other special populations, particularly at the county level. We propose a Bayesian method for producing subnational population projections, including migration and accounting for college populations, by building on but modifying the UN approach. We illustrate our approach by applying it to the counties of Washington State and comparing the results with extant deterministic projections produced by Washington State demographers. Out-of-sample experiments show that our method gives accurate and well-calibrated forecasts and forecast intervals. In most cases, our intervals were narrower than the growth-based intervals issued by the state, particularly for shorter time horizons.