Population Estimation for Census Tracts Using Dynamic Models

Marie V. Bousfield, Department of Planning, Chicago

This article describes how empirical Bayesian techniques and dynamic models can be used to derive postcensal population estimates by race for small areas such as census tracts. Vital rates estimates and their stochastic properties are developed. Independent of these, population forecasts are developed based on the stage of succession of the racial group. The dynamic population estimator we propose is a weighted average of the vital rates estimate and the forecast. The theory is illustrated by deriving annual estimates of the number of African American women, age 15 to 44, for the census tracts in Chicago for the 1980-90 period, using the 1980 census counts and annual counts of vital events for 1980-90. A comparison between the 1990 estimates and the 1990 census counts is made.