Bioequivalence and Bioavailability Forum • Potvin revis(it)ed

Potvin revis(it)ed [Two-Stage / GS Designs]

posted by Helmut – Vienna, Austria, 2014-06-03 00:07 (3973 d 16:32 ago) – Posting: # 13028
Views: 12,273

Dear Detlew!

Below the results of my simulations (I used α 0.0304 for “Method B” and 0.0282 for “Method C”). Power2Stage is an amazing piece – 73 minutes for Methods B and C! It took the PQRI 1½ years to come up with their simulations in Compaq Visual Fortran. ;-)

Method B: Empiric type I error; Pot = Potvin’s 0.0294, HB = homebrew’s 0.0304.

12 24 36 48 60 CV Pot HB Pot HB Pot HB Pot HB Pot HB 0.1 0.0297 0.0304 0.0294 0.0302 0.0294 0.0303 0.0292 0.0303 0.0294 0.0302 0.2 0.0463 0.0474 0.0320 0.0324 0.0294 0.0303 0.0292 0.0303 0.0297 0.0302 0.3 0.0437 0.0453 0.0475 0.0488 0.0397 0.0405 0.0324 0.0328 0.0296 0.0304 0.4 0.0344 0.0358 0.0433 0.0448 0.0485 0.0501 0.0458 0.0468 0.0409 0.0416 0.5 0.0309 0.0321 0.0338 0.0353 0.0420 0.0434 0.0484 0.0500 0.0483 0.0494 0.6 0.0297 0.0310 0.0307 0.0320 0.0333 0.0343 0.0399 0.0419 0.0466 0.0485 0.7 0.0294 0.0304 0.0299 0.0315 0.0306 0.0316 0.0328 0.0337 0.0381 0.0397 0.8 0.0292 0.0302 0.0298 0.0310 0.0303 0.0309 0.0303 0.0313 0.0318 0.0330 0.9 0.0289 0.0300 0.0298 0.0309 0.0296 0.0309 0.0297 0.0308 0.0300 0.0311 1.0 0.0291 0.0300 0.0298 0.0308 0.0298 0.0307 0.0297 0.0307 0.0301 0.0307

Method C: Empiric type I error; Pot = Potvin’s 0.0294, HB = homebrew’s 0.0282.

12 24 36 48 60 CV Pot HB Pot HB Pot HB Pot HB Pot HB 0.1 0.0496 0.0499 0.0500 0.0496 0.0500 0.0499 0.0501 0.0498 0.0504 0.0497 0.2 0.0510 0.0500 0.0490 0.0491 0.0499 0.0499 0.0495 0.0498 0.0500 0.0497 0.3 0.0441 0.0421 0.0492 0.0475 0.0477 0.0471 0.0494 0.0492 0.0502 0.0496 0.4 0.0346 0.0330 0.0435 0.0410 0.0489 0.0469 0.0469 0.0457 0.0470 0.0457 0.5 0.0311 0.0299 0.0339 0.0322 0.0418 0.0395 0.0480 0.0458 0.0483 0.0461 0.6 0.0299 0.0288 0.0307 0.0297 0.0331 0.0314 0.0399 0.0379 0.0472 0.0445 0.7 0.0294 0.0280 0.0298 0.0290 0.0308 0.0291 0.0325 0.0310 0.0380 0.0357 0.8 0.0292 0.0279 0.0301 0.0287 0.0299 0.0287 0.0302 0.0291 0.0319 0.0304 0.9 0.0285 0.0279 0.0298 0.0286 0.0296 0.0284 0.0298 0.0285 0.0301 0.0288 1.0 0.0290 0.0276 0.0295 0.0287 0.0297 0.0285 0.0297 0.0285 0.0297 0.0284

Maximum inflation in red.
With the new alphas no (!) significant inflation for both methods. Largest observed in “Method B” 0.050111 (at 36/0.4) and in “Method C” 0.049984 (at 12/0.2).

I’m getting the impression that if PQRI would have had a closer look right from the start (instead of coming up with a ‘one size fits all’ α and playing with a “negligible inflation”), maybe we could have avoided all those effectless discussions we had the last years. :crying:

BTW, in Montague’s paper I read “The simulations were performed using R […].” Nice to know. Then “A different randomly selected seed was used for each scenario.” Why? Shall we switch to setseed=FALSE in Power2Stage?
Anders’ algo suggests 0.027 (instead of 0.028) for “Method D”. Sim’s running.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

How to find a suitable adjusted α? Helmut 2014-05-31 18:20
- Suitable code for suitable adjusted α d_labes 2014-06-02 08:28
  - Handling inflation Helmut 2014-06-02 14:44
  - Potvin revis(it)edHelmut 2014-06-02 22:07
    - Potvin revis(it)ed d_labes 2014-06-03 08:47
      - Montague revis(it)ed Helmut 2014-06-03 13:49
      - Pocock’s “natural constant” Helmut 2014-10-13 14:53
        
        Pocock’s “natural constant” ElMaestro 2014-10-13 15:30
        Another “natural constant”? d_labes 2014-10-14 08:56
        
        Λ Helmut 2014-10-14 13:36