Table 2.

Model performances on forecasting hospitalization rates 7 days ahead in terms of point forecast metrics. Models were evaluated on 68 testing windows using the MAE and MAPE. Here the mean MAE and mean MAPE are displayed with the corresponding 95% confidence interval. Models were trained and tested autoregressive (AR) yes or no

Model	AR	MAE	MAPE in %
Linear Regression	X	0.54; (0.37, 0.70)	8.47; (7.07, 9.88)
		0.58; (0.33, 0.83)	8.50; (6.43, 10.56)
Ridge Regression	X	0.52; (0.38, 0.66)	8.05; (6.77, 9.34)
		0.52; (0.38, 0.65)	8.05; (6.77, 9.34)
ARIMA	X	0.29; (0.21, 0.37)	4.76; (3.79, 5.72)
		0.30; (0.21, 0.38)	5.42; (4.15, 6.70)
Random Forest	X	0.65; (0.38, 0.92)	8.54; (7.00, 10.08)
		0.44; (0.28, 0.61)	6.62; (5.29, 7.95)
XGBoost	X	0.66; (0.41, 0.90)	8.68; (7.09, 10.27)
		0.48; (0.32, 0.64)	7.33; (6.01, 8.65)