County-level homeless estimates, 2007-2025 (pre-imputation, with missing values)
Source:R/homeless_na.R
homeless_na.RdIdentical to homeless but before imputation: counties
that overlapped no HUD Continuum of Care in a given year retain a missing
(NA) count for that year. This is the input to the spatial-Poisson
imputation step that produces homeless.
Format
A data frame with 3143 rows and the same columns as
homeless; the yearly count07–count25 columns
contain NA for counties with no CoC coverage that year (about 1–3
percent of counties).
Source
HUD Point-in-Time data (see hud2007) disaggregated to
counties; ACS covariates via tidycensus.