Skip to main content

Table 3 Simulation study results for the Weibull spatial scan statistic

From: A log-Weibull spatial scan statistic for time to event data

Data distribution

IC

Power

PI

LC

TCa

TCc

M

V

a

b

c

a

b

c

a

b

c

a

b

c

a

b

c

Exponential

10

100.0

0.954

0.962

0.871

0.052

0.076

0.015

0.345

0.320

0.175

0.456

0.449

0.427

0.483

0.479

0.487

15

225.0

0.838

0.894

0.881

0.001

0.094

0.011

0.494

0.547

0.276

0.403

0.403

0.431

0.476

0.462

0.485

20

400.0

0.971

0.781

0.981

0.001

0.094

0.014

0.562

0.682

0.306

0.465

0.346

0.478

0.481

0.446

0.489

Weibull

10

4.0

0.732

0.701

0.755

0.538

0.961

0.976

0.000

0.006

0.000

0.306

0.240

0.258

0.461

0.327

0.329

15

10.0

0.697

0.973

0.704

0.869

0.317

0.993

0.131

0.672

0.000

0.243

0.378

0.238

0.308

0.407

0.303

20

7.0

0.806

0.993

0.715

0.652

0.879

0.966

0.172

0.121

0.034

0.304

0.340

0.245

0.418

0.349

0.307

Log-Normal

10

4.0

0.672

0.726

0.427

0.074

0.176

0.027

0.248

0.824

0.156

0.315

0.283

0.211

0.458

0.363

0.459

15

10.0

0.721

0.971

0.599

0.000

0.221

0.043

1.000

0.779

0.957

0.290

0.374

0.240

0.372

0.388

0.352

20

17.0

0.256

0.999

0.835

0.164

0.287

0.309

0.836

0.713

0.691

0.105

0.380

0.320

0.248

0.387

0.371

Gamma

10

5.0

0.373

0.230

0.355

0.048

0.263

0.062

0.584

0.737

0.214

0.163

0.090

0.171

0.398

0.226

0.440

15

7.5

0.401

0.517

0.405

0.181

0.000

0.018

0.819

1.000

0.664

0.158

0.210

0.175

0.300

0.339

0.403

20

10.0

0.443

0.713

0.406

0.173

0.000

0.093

0.826

1.000

0.906

0.176

0.289

0.161

0.312

0.371

0.306

Log-Weibull

10

5.5

0.672

0.298

0.654

0.282

0.059

0.022

0.054

0.553

0.091

0.308

0.138

0.323

0.458

0.385

0.472

15

6.0

0.717

0.344

0.688

0.005

0.192

0.000

0.031

0.754

0.080

0.360

0.150

0.343

0.482

0.387

0.478

20

6.5

0.668

0.309

0.716

0.138

0.185

0.001

0.518

0.764

0.002

0.297

0.135

0.362

0.443

0.377

0.484

  1. Five probability models each with three different means inside true cluster are used under three right censoring cases: a = 20%:20%, b = 20%:40%, c = 40%:20% outside cluster: mean = 2; variance = 4(Exponential), 0.188(Weibull), 2(log-Normal), 1(Gamma), and 5(log-Weibull). IC inside cluster, M mean, V variance, PI perfect identification, LC large cluster, identification TC a  average Tanimoto coefficient, TC c  cumulated Tanimoto coefficient