Spatio-racial distributions in major US cities change on the timescale of a single decade. Here we describe a methodology to forecast such changes a decade ahead. First, we transform the data from population counts to a grid of categorical population types. Then, we build an empirical model of past change using supervised machine learning and extrapolate it into the future to make a prediction. The model uses only statistics of population categories as features, there are no ancillary variables. To account for the non-stationarity of the change we use a synthetic training dataset based on past transitions and estimated future frequencies of these transitions. The methodology is described and validated by training a model on 1990-2000 data and using it to predict spatio-racial distributions in 2010. This prediction is then compared to the actual spatio-racial 2010 distribution. We have found that a highly accurate model of change can be constructed using this methodology. Extrapolating such models to the future results in some loss of accuracy, but the method still yields satisfactory predictions.

Racial geography in US urban areas is extensively studied with the emphasis on assessing the extent of racial segregation. However, the used methodology has not changed for at least two decades; it relies on calculating ratios of population counts in the entire city and its subdivisions – census aggregation areas. This has a number of limitations; the two most important are: assessment of segregation depends on the subdivisions used, segregation can only be calculated for regions with census subdivisions. Here we present a different conceptualization of racial geography, which leads to a new method called racial landscape (RL). We use block-level census data to construct a high-resolution grid where each cell represents single race inhabitants. The result is a spatial, racial pattern; a degree of spatial autocorrelation of this pattern is a measure of segregation that does not require using subdivisions. We shortly describe the RL method and its application to Cook County, IL. We also describe here its implementation in the R computational environment.