Empirical Prediction Intervals for County Population Forecasts

Stefan Rayer, University of Florida
Stanley K. Smith, University of Florida
Jeff Tayman, San Diego Association of Governments (SANDAG)

Population forecasts entail a significant amount of uncertainty, especially for long-range horizons and for places with small or rapidly changing populations. This uncertainty can be dealt with by presenting a range of projections or by developing statistical prediction intervals based on models that incorporate the stochastic nature of the forecasting process or on empirical analyses of past forecast errors. In this paper, we develop and test empirical prediction intervals for county population forecasts in the United States. We find that prediction intervals based on the distribution of past forecast errors provide reasonably accurate predictions of the distribution of future forecast errors. We believe the construction of empirical prediction intervals to accompany population forecasts will help data users plan more effectively for an uncertain future.