Table 2.

Performance metrics of the different models used. For the original data a linear multiple regression model was used (Anegagrie et al., 2021), and then the baseline (which included all the variables except the number of people per house) and final XGBoost model (including the top five variables and the number of people per household) for the current study. In adition: R² train refers to R-squared of the training data set, MSE train refers to the Mean Square Error of the test data set, R² test refers to R-squared of the training data set, MSE test refers to the Mean Square Error of the test data set.

	R2 train	MSE train	R2 test	MSE test
Linear Multiple regression model	0.76	0.68	0.71	0.64
Baseline XGBoost model	0.37	1.72	−0.10	1.94
Final XGBoost model	0.99	0.01	0.79	0.36