Table 5.
Detection power with and without randomization testing
| method | ED dataset | TH dataset | CC dataset | AF dataset |
| KULL MA | 3.23/3.23 | 4.41/4.23 | 5.40/2.55 | 4.52/3.19 |
| KULL MA-DOW | 3.45/3.04 | 4.95/4.09 | 3.73/2.26 | 3.65/2.80 |
| KULL MA-WK | 3.31/3.23 | 5.26/4.23 | 6.04/2.55 | 5.30/3.19 |
| KULL MA-WK-DOW | 3.19/3.04 | 5.20/4.09 | 3.57/2.26 | 3.99/2.80 |
| EBP MA | 2.54/2.50 | 3.95/3.29 | 6.36/4.16 | 5.89/3.99 |
| EBP MA-DOW | 2.65/2.53 | 3.51/3.44 | 4.59/4.10 | 5.62/4.36 |
| EBP MA-WK | 2.74/2.50 | 5.04/3.40 | 5.84/2.70 | 5.11/3.92 |
| EBP MA-WK-DOW | 2.92/2.59 | 4.31/3.75 | 5.05/2.47 | 5.30/4.00 |
| EBG MA | 4.50/2.91 | 5.90/4.19 | 4.94/4.43 | 4.92/4.63 |
| EBG MA-DOW | 5.48/3.01 | 5.15/4.66 | 5.61/4.50 | 5.00/4.79 |
| EBG MA-WK | 4.87/2.87 | 5.92/4.24 | 3.82/3.16 | 4.58/4.43 |
| EBG MA-WK-DOW | 5.53/3.04 | 5.92/4.73 | 4.90/2.96 | 4.53/4.56 |
Average days to detection at 1 false positive per month for "medium-sized" outbreaks injected into each dataset, using empirically determined thresholds on p-value (computed by randomization testing) and score (without randomization testing) respectively. If there is a significant difference between the detection times with and without randomization, the better-performing method is marked in bold.