I am currently working on analyzing the behavior of two algorithms on the same input data and am trying to establish whether they are reasonably equivalent to each other within specified bounds to this end I am attempting to use the sampleN.TOST from the following r package.
To calculate the needed sample size for this analysis.
What I am failing to understand is the following:
How should the inputs be formatted to produce consistent estimation i.e. how should I be thinking about calculating the equivalence limits etc.
How is the CV value arrived at. Is it from a prior data or is there a method for selecting one in absence of such data.
Pardon in advance if this is obvious or on the wrong SE.