harmonic mean for half-life?

[slyrus]

It's great that I can use median, geomean, etc... with pknca parameters. But this leads me to 1) should I be using the harmonic mean for t1/2? and 2) if so, should PKNCA have a harmmean function analogous to the geomean function?

[billdenney]

For the first question, I agree that the harmonic mean is a more theoretically-sound choice for central tendency. But for the second question, I’m hoping to move functions that aren’t core to PKNCA out of the package (see #71).