Freedom for replicate bears [R for BE/BA]

posted by ElMaestro  – Denmark, 2009-04-17 14:50 (4569 d 10:10 ago) – Posting: # 3556
Views: 64,520

Hi dlabes,

» Or is the denominator DF question also like type III SumOfSquares? (I have read some Websites indicating that).

As far as I recall, the R default -with no option to change it- is to use a denDF which is similar to SAS' "containment" meffud. I you insist on having Satterthwaite's denDF or whatever, then R will not be your package of choice. One little problem is that FDA indirectly endorsed SAS some time ago, so many regulators (incl. EU) expect an output which can be replicated with SAS using FDA scripts. Then clearly R will fall short of the goal. I guess you can safely conclude this is another "type III SS" or LSMean phenomenon. Big bro dictates the standard and regulators have no clue. Do it the SAS way or do it wrong. OK, that was a provocation, sorry.

One can go on and on; discussion of the proof that REML gives unbiased
coefficients in unbalanced datasets comes next, I guess?

Best regards

Complete thread:

 Admin contact
21,749 posts in 4,547 threads, 1,545 registered users;
online 6 (0 registered, 6 guests [including 4 identified bots]).
Forum time: Thursday 01:01 CEST (Europe/Vienna)

Imagine if every Thursday your shoes exploded
if you tied them the usual way.
This happens to us all the time with computers,
and nobody thinks of complaining.    Jef Raskin

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