Longitudinal analysis reveals that simple naive models match complex population forecast accuracy over five decades, highlighting the need for systematic benchmark evaluations.
Highlights This study is the first in the San Diego region to evaluate population forecast error across multiple forecast vintages and increment years spanning more than five decades. Forecast accuracy declines with increasing forecast horizon, while forecast errors are also influenced by population growth rate, population size, and jump-off discrepancy. Findings confirm the generalizability of established relationships in the forecast evaluation literature while extending prior research through analysis of errors in population change rather than solely through population level. Simple naive extrapolation methods frequently matched or exceeded the accuracy of more complex forecasting approaches, demonstrating the importance of benchmarking forecast performance.
No takes yet. Share an insight, caveat, or question.
Schroeder et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: