Table 1.

Correlation between the three evaluation indices (sensitivity, specificity and time to detection) and parameter values for five algorithms with P values^#.

Algorithms	Sensitivity		Specificity		Time to detection
	Pearson’s r	P value	Pearson’s r	P value	Pearson’s r	P value
λ value for EWMA (k = 1.0)	—	—	0.526	0.118	−0.897	0.000⁎
λ value for EWMA(k = 2.0)	0.837	0.003⁎	0.684	0.029⁎	−0.990	0.000⁎
λ value for EWMA (k = 3.0)	0.921	0.000⁎	0.780	0.008⁎	−0.990	0.000⁎
k value for C1′ (H = 3σ)	−0.918	0.000⁎	0.969	0.000⁎	0.987	0.000⁎
k value for C1′ (H = 4σ)	−0.973	0.000⁎	0.937	0.000⁎	0.992	0.000⁎
k value for C1′ (H = 5σ)	−0.908	0.000⁎	0.904	0.001⁎	0.995	0.000⁎
k value for C2′ (H = 3σ)	−0.759	0.000⁎	0.982	0.000⁎	0.973	0.000⁎
k value for C2′ (H = 4σ)	−0.907	0.000⁎	0.961	0.000⁎	0.984	0.000⁎
k value for C2′ (H = 5σ)	−0.938	0.000⁎	0.943	0.001⁎	0.959	0.000⁎
k value for C3′ (H = 3σ)	−0.694	0.004⁎	0.998	0.000⁎	0.956	0.000⁎
k value for C3′ (H = 4σ)	−0.818	0.000⁎	0.994	0.000⁎	0.962	0.000⁎
k value for C3′ (H = 5σ)	−0.866	0.000⁎	0.995	0.000⁎	0.960	0.000⁎
Spatial cluster size for space–time permutation scan statistic (3d)	−0.958	0.042⁎	—	—	0.962	0.038⁎
Spatial cluster size for space–time permutation scan statistic (7d)	−0.835	0.165	—	—	0.756	0.244
Spatial cluster size for space–time permutation scan statistic (10d)	−0.746	0.254	—	—	0.538	0.462

^#Pearson’s r was calculated directly for those variables showing linear relation. While for those variables showing relation of log linear, Pearson’s r was calculated after applying logarithmic transformation.

^⁎P value was less than 0.05 (2-tailed).