Spatial cluster detection using dynamic programming

Table 1 Model averaging comparison in terms of area under the ROC curve

Frequency range	Coverage	DP AUC	SR AUC	(DP-SR) 95% confidence interval limits		p-value
Low	< 16%	0.95047	0.95016	- 0.00746	0.00808	0.9384
	16-34%	0.86260	0.86852	- 0.01261	0.00078	0.0831
	≥ 35%	0.97409	0.97911	- 0.01005	0.00001	0.0505
Mid	< 16%	0.96019	0.96079	- 0.00530	0.00410	0.8014
	16-34%	0.98246	0.98613	- 0.00575	- 0.00159	0.0005
	≥ 35%	1.00000	1.00000
High	< 16%	0.98468	0.98379	- 0.00173	0.00352	0.5044
	16-34%	0.99990	0.99984	- 0.00001	0.00013	0.1096
	≥ 35%	1.00000	1.00000

Comparison of the DP and SR ROC curves obtained from model averaging. The table shows the AUC for each algorithm, the lower and upper 95% confidence limits for the difference in areas (DP-SR) and the associated p-values

ISSN: 1472-6947