Table 2.

Assessment of the discriminating power of BRT models for malaria risk in villages.

	Model	Overall		P. vivax		P. falciparum
		cvAUC	tAUC	cvAUC	tAUC	cvAUC	tAUC
High risk (API > 10)	2010	0.72	0.70	0.72	0.70	0.78	0.76
	2011	0.80	0.76	0.80	0.75	0.86	0.74
	2012	0.82	0.80	0.82	0.80	0.84	0.81
	2013	0.84	0.80	0.83	0.80	0.87	0.80
	2014	0.83	0.82	0.82	0.82	0.85	0.84
	2015	0.82	0.79	0.82	0.79	0.87	0.85
	2016	0.82	0.80	0.82	0.81	0.87	0.84
	2017	0.82	—	0.83	—	0.87	—
Very high risk (API > 50)	2010	0.76	0.76	0.77	0.76	0.82	0.78
	2011	0.85	0.81	0.84	0.80	0.88	0.81
	2012	0.85	0.82	0.85	0.81	0.87	0.76
	2013	0.86	0.80	0.86	0.80	0.89	0.84
	2014	0.84	0.84	0.84	0.83	0.89	0.83
	2015	0.85	0.83	0.85	0.82	0.89	0.86
	2016	0.84	0.83	0.84	0.83	0.88	0.83
	2017	0.85	—	0.84	—	0.89	—

Each cross-validation BRT model built with data of a given year yielded a cross-validated AUC (cvAUC), while its model predictions with testing data of the following year allowed for the estimation of a testing AUC (tAUC).