TIE, repeat once more please... [RSABE / ABEL]

posted by Astea – Russia, 2020-04-02 12:41 (129 d 03:06 ago) – Posting: # 21313
Views: 3,842

Dear Friends!

I'm very sorry for digging this theme once more, but my female logic suffers from trying to get how could the sequence of statements influence the TIE. I suppose the TIE in alternative Elena's approach to be the same as in the standard scheme. Would you be so kind to explain it for one thousand and first time again?

Let A be the statement that CI is in [80,125], B - the statement that CV<30% and C - the statement that CI is [L,U], where L and U are the scaled limits. Then BE statement is
(B∧A)∨(¬B∧C) for standard approach and (A∨(¬A∧(¬B∧C))) for the alternative approach. The non BE statements are (¬A∧B)∨(¬A∧¬C)∨(¬B∧¬C) and (¬A∧B)∨(¬A∧¬C) correspondingly. The only difference is (¬B∧¬C) which is logically equals to (¬A∧¬C) when ¬B is TRUE and zero if ¬B is FALSE.

Testing the data twice is obvious in the alternative sequence, but the second test doesn't suffer from the error on CV_R because the first test A has already done. And in the standard approach we also assess the data twice: first to get CV_R, then to calculate CI. I'm totally lost: my gut feeling says that there should not be difference, TIE is the same in both cases. I read the thread once and once again and still have internal contradictions that I couldn't prove.

Practically: while dealing with scaled approach I program to get CI as well as CV_R in the output at the same time. What I should do - close the left eye not to see that CI is in the [80-125] and only looking for the CV_R by my right eye?

"Being in minority, even a minority of one, did not make you mad"

Complete thread:

Activity
 Admin contact
20,986 posts in 4,375 threads, 1,460 registered users;
online 16 (1 registered, 15 guests [including 11 identified bots]).
Forum time: Sunday 15:48 CEST (Europe/Vienna)

The interpretation of facts in a certain way
stimulates other scientists’ thoughts.    Róbert Bárány

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