Fig. 1From: A machine learning approach to small area estimation: predicting the health, housing and well-being of the population of NetherlandsIllustration of small area estimation at neighbourhood level in the Netherlands. Prevalence of “drank alcohol in the past 12 months” based on survey responses (raw estimates), XGBoost model predictions for the population (model estimates), and the percentage point difference of two models (XGBoost vs. STAR). XGBoost is based on X and Y-coordinatesBack to article page