Estimating the Civilian Noninstitutional Population for Small Areas: A Modified Cohort Component Approach Using Public Use Data

Abstract

This article develops a demographic method to estimate the civilian noninstitu- tional population for counties and county equivalents in the U.S. While these data provide the key sampling frame for national labor market surveys and denominators for labor market prevalence rates, the data are thus far unavailable for small areas. I develop a modified cohort component method to produce novel, monthly estimates of the civilian noninstitutional population for all U.S. counties using publicly avail- able data on population and vital statistics with minimal modifications. The resulting population data may be used by researchers and policymakers to study within-year population dynamics as they relate to economic and demographic factors. I further extend the method to produce short-term population projections that include the most current vital statistics. The method compares favorably to existing annual, midyear estimates by the U.S. Census Bureau, but is prone to error in areas with fewer vital events.

Andrew C. Forrester

Economist and Statistician in Washington, DC

Economist and statistician in Washington, DC working on economic statistics, labor and financial economics, time series and seasonal adjustment, and quantitative demography. All views are my own.