Skip to main content

Table 2 Power analysis for the overall test and subgroup tests of the five binary classifiers for eight scenarios (A-H). The overall power was calculated as the original reference [11], which is the sum of the number of overall test < 0.02 and the number of significance in comparison 2 of overall test > 0.02

From: Subgroup identification for treatment selection in biomarker adaptive design

Scenario A B C D E F G H
Overall Test < 0.02 64 70 69 924 159 997 463 487
Comparison 2_DLDA 26 35 33 483 153 503 455 481
ASD (1,1) 33 34 32 822 115 951 436 455
ASD (1,2) 20 25 25 682 116 859 451 476
ASD (2,1) 24 33 33 587 146 665 350 392
ASD (2,2) 23 33 31 488 153 411 286 352
Overall Test > 0.02 936 930 931 76 841 3 537 513
Comparison 2_DLDA 319 355 401 24 771 2 510 488
ASD (1,1) 130 131 115 11 298 0 388 361
ASD (1,2) 247 266 308 18 676 2 492 475
ASD (2,1) 216 260 280 20 709 2 226 237
ASD (2,2) 174 219 255 23 717 1 142 160
Overall power_DLDA 383 425 470 948 930 999 973 975
ASD (1,1) 194 201 184 935 457 997 851 848
ASD (1,2) 311 336 377 942 835 999 955 962
ASD (2,1) 280 330 349 944 868 999 689 724
ASD (2,2) 238 289 324 947 876 998 605 647