Skip to main content

Table 5 Different accuracy metrics for “drinker” indicator

From: A machine learning approach to small area estimation: predicting the health, housing and well-being of the population of Netherlands

Model Accuracy AUC MSE NLL Time
Null 0.801 0.500 0.160 0.500 0 s
STAR 0.814 0.737 0.138 0.437 28 min
XGBoost0_xy 0.815 0.742 0.137 0.434 6s
XGBoost_xy 0.816 0.744 0.136 0.433 3 min
XGBoost0_ogc 0.815 0.742 0.137 0.434 17s
XGBoost_ogc 0.816 0.744 0.136 0.433 6 min
  1. Bold values indicated the best result, not statistical significance