Table 2.
The best selected autoregressive integrated moving average (ARIMA) models using the Bayesian information criterion (BIC) and Akaike information criterion (AIC).
| Variable | ARIMA | Log likelihood | AIC | BIC |
| Cycle threshold value | (1,0,1)a | –355.99 | 702.38 | 716.42 |
| Number of hospitalized patients | (1,0,1) | –1553.31 | –1231.24 | –1220.61 |
| Number of positive COVID-19 tests | (1,0,1) | –1198.16 | 2494.99 | 2827.07 |
| Number of COVID-19 deaths | (1,0,2) | –933.16 | 1876.33 | 1893.88 |
| Number of hospitalized patients under 5 years old | (1,0,1) | –905.64 | 481.26 | 494.797 |
| Number of hospitalized patients aged 5-17 years | (1,0,1) | –819.80 | 393.80 | 407.322 |
| Number of hospitalized patients aged 18-59 years | (1,0,1) | –1401.00 | 374.58 | 384.72 |
| Number of hospitalized patients over 60 years old | (1,0,1) | –919.60 | 397.41 | 407.55 |
aThe numbers in parentheses represent the parameters (p, d, q) of the model, where p is the order of autoregression, d is the degree of trend difference, and q is the order of moving average.