Optimizing the maximum reported cluster size for the multinomial-based spatial scan statistic

Moon, Jisu; Kim, Minseok; Jung, Inkyung

doi:10.1186/s12942-023-00353-4

International Journal of Health Geographics

Table 3 Multinomial model: simulation results for the true cluster model (B) and alternative hypothesis (2) using elliptical windows

From: Optimizing the maximum reported cluster size for the multinomial-based spatial scan statistic

		Maximum reported cluster size (MRCS)																		Default setting
		1%	2%	3%	4%	5%	6%	8%	10%	12%	15%	20%	25%	30%	35%	40%	45%	50%	Overall	Default setting
SCIC₁	Freq^a	0	2	4	11	43	54	403	132	72	53	25	12	7	3	1	1	3	826	826
	Sen^b	NA	0.100	0.350	0.400	0.567	0.641	0.890	0.923	0.961	0.970	0.840	0.817	1.000	0.933	1.000	1.000	1.000	0.862	0.876
	PPV^c	NA	0.500	1.000	0.879	0.927	0.894	0.959	0.763	0.651	0.545	0.348	0.256	0.275	0.203	0.185	0.172	0.139	0.824	0.752
	Mis^d	NA	0.072	0.047	0.049	0.035	0.032	0.011	0.028	0.043	0.065	0.130	0.200	0.193	0.271	0.319	0.348	0.469	0.035	0.060
SCIC₂	Freq	0	2	3	10	40	53	392	130	68	51	26	15	11	8	4	5	8	826	826
	Sen	NA	0.100	0.333	0.380	0.585	0.649	0.890	0.932	0.956	0.969	0.838	0.853	0.945	0.825	0.800	0.960	0.975	0.865	0.876
	PPV	NA	0.500	1.000	0.867	0.938	0.910	0.960	0.772	0.646	0.544	0.340	0.240	0.244	0.155	0.131	0.146	0.118	0.805	0.752
	Mis	NA	0.072	0.048	0.051	0.034	0.031	0.011	0.027	0.044	0.066	0.134	0.220	0.223	0.346	0.402	0.414	0.567	0.046	0.060
Elbow	Freq	0	1	3	11	51	57	412	129	70	41	20	13	8	4	3	1	0	824	826
	Sen	NA	0.000	0.333	0.509	0.600	0.653	0.891	0.927	0.954	0.966	0.800	0.877	0.850	0.900	0.467	1.000	NA	0.859	0.876
	PPV	NA	0.000	1.000	0.879	0.913	0.908	0.955	0.761	0.648	0.539	0.312	0.240	0.219	0.164	0.083	0.161	NA	0.828	0.752
	Mis	NA	0.087	0.048	0.041	0.034	0.031	0.012	0.029	0.044	0.067	0.147	0.224	0.250	0.355	0.357	0.377	NA	0.035	0.060
MCS-P	Freq	0	2	5	11	46	51	378	121	74	60	36	18	11	3	5	3	2	826	826
	Sen	NA	0.000	0.280	0.473	0.565	0.655	0.886	0.924	0.962	0.967	0.878	0.911	0.855	0.933	0.960	1.000	1.000	0.862	0.876
	PPV	NA	0.000	0.800	0.788	0.870	0.899	0.961	0.775	0.671	0.551	0.380	0.302	0.235	0.203	0.187	0.167	0.155	0.802	0.752
	Mis	NA	0.087	0.058	0.046	0.038	0.031	0.011	0.025	0.037	0.062	0.115	0.159	0.213	0.271	0.304	0.362	0.399	0.038	0.060
MCHS-P	Freq	0	2	4	9	47	50	376	123	76	60	35	19	12	3	5	3	2	826	826
	Sen	NA	0.000	0.350	0.556	0.591	0.652	0.887	0.922	0.961	0.970	0.886	0.884	0.867	0.933	0.960	1.000	1.000	0.866	0.876
	PPV	NA	0.000	1.000	0.807	0.891	0.903	0.961	0.767	0.662	0.548	0.383	0.287	0.234	0.203	0.187	0.167	0.155	0.800	0.752
	Mis	NA	0.087	0.047	0.042	0.036	0.031	0.011	0.027	0.040	0.063	0.114	0.175	0.217	0.271	0.304	0.362	0.399	0.039	0.060

^aFreq: frequency
^bSen: sensitivity
^cPPV: positive predictive value
^dMis: misclassification
Sensitivity, PPV, and misclassification rate at the most frequently selected optimal MRCS are shown in bold

Back to article page

ISSN: 1476-072X

Contact us

Submission enquiries: journalsubmissions@springernature.com