Research Article
Detecting Spatial Clusters via a Mixture of Dirichlet Processes
Table 1
Simulation assessments for three foci-level clusters.
| Scenario () | Median Num. of Clusters (SD) | Cluster index | Average Sensitivity (SD) | Average Specificity (SD) | Average Accuracy (%) (SD) |
| Normal (0.1, 1.0) | IC: 3 (0.88) | 1 | 0.96 (0.13) | 1 (0) | 0.96 (0.13) | 2 | 0.96 (0.13) | 1 (0) | 3 | 0.96 (0.13) | 1 (0) | SC: 2 (0.32) | 1 | 0.93 (0.26) | 1 (0.04) | 0.96 (0.13) | 2 | 1 (0.04) | 0.93 (0.26) |
| Chi-squared (0.05, 0.1) | IC: 19 (2.17) | 1 | 0.94 (0.09) | 0.91 (0.15) | | 2 | 0.74 (0.43) | 0.96 (0.02) | 0.54 (0.13) | 3 | 0.09 (0.03) | 0.94 (0.16) | SC: 2 (0.56) | 1 | 0.76 (0.37) | 0.9 (0.1) | 0.82 (0.18) | 2 | 0.88 (0.11) | 0.78 (0.36) |
| Large 1 (0.05, 0.1) | IC: 3 (0.82) | 1 | 0.96 (0.13) | 1 (0) | 0.96 (0.13) | 2 | 0.96 (0.13) | 1 (0) | 3 | 0.96 (0.13) | 1 (0) | SC: 2 (0.29) | 1 | 0.92 (0.27) | 1 (0) | 0.96 (0.14) | 2 | 1 (0) | 0.92 (0.27) |
| Large 2 (0.01, 0.5) | IC: 3 (0.77) | 1 | 0.99 (0.04) | 1 (0) | 0.99 (0.02) | 2 | 0.99 (0.03) | 1 (0) | 3 | 1 (0.01) | 1 (0) | SC: 2 (0.14) | 1 | 1 (0.02) | 1 (0.01) | 1 (0.01) | 2 | 1 (0.01) | 1 (0) |
| Large 3 (0.05, 0.05) | IC: 5 (1.25) | 1 | 0.91 (0.13) | 1 (0) | 0.92 (0.07) | 2 | 0.88 (0.16) | 1 (0) | 3 | 0.96 (0.09) | 1 (0) | SC: 2 (0.37) | 1 | 0.97 (0.06) | 0.98 (0.03) | 0.97 (0.04) | 2 | 0.98 (0.03) | 0.98 (0.03) |
| Large 4 (0.01, 2.0) | IC: 9 (1.3) | 1 | 0.66 (0.14) | 0.99 (0.01) | 0.62 (0.09) | 2 | 0.33 (0.08) | 0.99 (0.01) | 3 | 0.8 (0.17) | 1 (0) | SC: 1 (0.36) | 1 | 0.04 (0.17) | 0.99 (0.05) | 0.51 (0.08) | 2 | 0.98 (0.07) | 0.05 (0.18) |
|
|
SD: standard deviation across 100 MC replicates; IC: individual foci cluster; SC: study effect clusters. |