Skip to main content
. 2022 Jul 14;10(7):1310. doi: 10.3390/healthcare10071310

Table 3.

The model training diagnostic results on the COVID-19 data performed by candidate models.

Model AIC RMSE MAPE Ljung–Box Test
Score p-Value
Confirmed cases
SARIMA(1, 1, 1),(0, 1, 1, 7) 1336.32 1931.46 7.26% 0.14 0.71
SARIMA(1, 1, 1),(0, 1, 2, 7) 1336.41 2230.31 8.18% 0.18 0.68
SARIMA(1, 1, 1),(1, 1, 1, 7) 1336.42 2281.49 8.31% 0.15 0.7
SARIMA(0, 1, 1),(1, 1, 1, 7) 1338.13 3185.02 11.28% 0.82 0.37
SARIMA(1, 1, 2),(0, 1, 1, 7) 1338.29 1999.01 7.44% 0.14 0.71
Deaths
SARIMA(0, 1, 2),(1, 1, 2, 7) 756.04 8.93 5.87% 0.07 0.8
SARIMA(0, 1, 2),(1, 1, 1, 7) 754.13 8.90 5.82% 0.05 0.82
SARIMA(1, 1, 1),(1, 1, 2, 7) 756.57 9.27 6.22% 0.18 0.68
SARIMA(1, 1, 1),(1, 1, 1, 7) 754.30 8.99 5.87% 0.1 0.76
SARIMA(1, 1, 2),(1, 1, 2, 7) 758.35 9.01 5.91% 0.07 0.79
Recovered cases
SARIMA(0, 1, 1),(0, 1, 1, 7) 1410.43 3686.54 13.51% 0.14 0.71
SARIMA(0, 1, 2),(0, 1, 1, 7) 1411.39 3799.29 13.90% 0.49 0.48
SARIMA(1, 1, 1),(0, 1, 1, 7) 1411.87 3750.77 13.74% 0.39 0.53
SARIMA(0, 1, 1),(1, 1, 1, 7) 1412.22 3312.99 12.13% 0.09 0.76
SARIMA(0, 1, 1),(0, 1, 2, 7) 1412.23 3343.85 12.25% 0.10 0.75