Table 1.
Maximum type 1 error rate for
with and without treatment selection, with or without adjustment for multiplicity and with equal or flexible second-to-first stage ratios as compared to the case 
| nominal α | ![]() |
![]() |
||||
|---|---|---|---|---|---|---|
![]() |
![]() |
![]() |
||||
| treatment selection of most promising treatment | ||||||
| equal (Proschan and Hunsberger, 1995) | flexible (Graf and Bauer, 2011) | equal (Section 4.1) | flexible (Section 4.2) | equal (Section 4.1) | flexible (Section 4.2) | |
| 0.01 | 0.0267 | 0.0491 | 0.0398 | 0.0697 | 0.0224 | 0.0407 |
| 0.025 | 0.0616 | 0.1064 | 0.0887 | 0.1466 | 0.0518 | 0.0892 |
| 0.05 | 0.1146 | 0.1867 | 0.1594 | 0.2496 | 0.0968 | 0.1588 |
| without treatment selection | ||||||
| equal (Proschan and Hunsberger, 1995) | flexible (Graf and Bauer, 2011) | equal (Section 5.1) | flexible (Section 5.2) | equal (Section 5.1) | flexible (Section 5.2) | |
| 0.01 | 0.0267 | 0.0491 | 0.0478 | 0.0800 | 0.0263 | 0.0473 |
| 0.025 | 0.0616 | 0.1064 | 0.1058 | 0.1701 | 0.0610 | 0.1037 |
| 0.05 | 0.1146 | 0.1867 | 0.1897 | 0.2885 | 0.1138 | 0.1842 |




