The ultimate crackpot iteration! [Two-Stage / GS Designs]

posted by ElMaestro  – Belgium?, 2017-08-20 17:32  – Posting: # 17728
Views: 24,325

Hi Hötzi,

we are not on the same page. Again :-D:-D
I am not sure I get it all, but I think with this code you are calling the same shit over and over again and that is completely unnecessary.
In the beginning of everything, before you initiate the simulation, literally before you make use of the nsims object, you read out the assumed GMR and target power, and on basis of that you can define an array of critical CV's which you initialise with the correct values. Once and for all. Not for every call where you need a sample size.

Note that if you initialise the array of critical values with syntax such as
vecCritCVs<<- rep(0, 400)

or however many values you need then vecCV will be public (I think), due to "<<-".
You can then access the array of critical CV's from within the SampleN.Stg2 function.

We might end up with something like:

sampleN.Stg2.EM <- function(CV)
{
  #all other variable info is already accumulated thin the critical values, so:
  Nps=1
  while (vecCritCVs[Nps]<=CV) Nps=Nps+1
  return(Nps*2) 
}


Can you see what I mean now?:cool::cool:

I could be wrong, but...

Best regards,
ElMaestro

"Pass or fail" (D. Potvin et al., 2008)

Complete thread:

Activity
 Admin contact
20,779 posts in 4,351 threads, 1,444 registered users;
online 34 (1 registered, 33 guests [including 20 identified bots]).
Forum time: 09:36 UTC (Europe/Vienna)

All we know about the world teaches us that the effects of A and B
are always different—in some decimal place—for any A and B.
Thus asking “are the effects different?” is foolish.    John W. Tukey

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5