Skip to main content

Table 1 Simulation scenarios for the true cluster model and alternative hypothesis

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

Setting

True cluster model

Alternative hypothesisa

Single cluster

(A) One circular-shaped cluster (8%)

(B) One elliptic-shaped cluster (8%)

(C) One irregular-shaped cluster (15%)

(1) p(1) = (0.05, 0.15, 0.35, 0.45)

(2) p(1) = (0.05, 0.25, 0.25, 0.45)

(3) p(1) = (0.10, 0.10, 0.40, 0.40)

(4) p(1) = (0.15, 0.15, 0.15, 0.55)

Two homogeneous clusters

(D) Two circular-shaped clusters (8% each)

(E) Two elliptic-shaped clusters (8% each)

(1) p(1) = p(2) = (0.05, 0.15, 0.35, 0.45)

(2) p(1) = p(2) = (0.05, 0.25, 0.25, 0.45)

(3) p(1) = p(2) = (0.10, 0.10, 0.40, 0.40)

(4) p(1) = p(2) = (0.15, 0.15, 0.15, 0.55)

Two heterogeneous clusters

(D) Two circular-shaped clusters (8% each)

(E) Two elliptic-shaped clusters (8% each)

(5) p(1) = (0.05, 0.15, 0.35, 0.45), p(2) = (0.05, 0.25, 0.25, 0.45)

(6) p(1) = (0.05, 0.15, 0.35, 0.45), p(2) = (0.10, 0.10, 0.40, 0.40)

(7) p(1)= (0.05, 0.15, 0.35, 0.45), p(2) = (0.15, 0.15, 0.15, 0.55)

  1. ap(1) is for cluster 1 and p(2) is for cluster 2; p(0) = (0.25, 0.25, 0.25, 0.25) was assumed for the remaining areas